Full text of "Archaic Roman Religion"

See other formats

ARCHAIC

ROMAN

RELIGION

WITH AN APPENDIX ON
THE RELIGION OF THE ETRUSCANS

GEORGES DUMEZIL

TRANSLATED BY PHILIP KRAPP

FOREWORD BY MIRCEA ELIADE

VOLUME ONE

THE JOHNS HOPKINS UNIVERSITY PRESS

BALTIMORE AND LONDON

Originally published in 1966 as

La Religion rotnaine archaique suivi d y t*n appendice sur la religion des Etrusques
© 1966, Editions Payot, Paris

The English translation has been revised and annotated by the author

Printed in the United States of America on acid-free paper
Originally published in the United States in 1970 as a two-volume set by the
University of Chicago Press

Licensed by the University of Chicago Press, Chicago, Illinois
© 1970 by The University of Chicago. All rights reserved

Johns Hopkins edition, 1996
05 04 03 02 01 00 99 98 97 96 54321

The Johns Hopkins University Press
2715 North Charles Street
Baltimore, Maryland 21218-4319
The Johns Hopkins Press Ltd., London

Library of Congress Cataloging-in-Publication Data
Dumdzil, Georges, 1898-
[Religion romaine archaique. English]

Archaic Roman religion : with an appendix on the religion of the Etruscans / Georges
Dumdzil : translated by Philip Krapp : foreword by Mircea Eliade.
p. cm.

Originally published : Chicago : University of Chicago Press, 1970.

Includes bibliographical references and index.

ISBN 0-8018-5482-2 (v. 1. : alk. paper).— ISBN 0-8018-5480-6 (v. 1. : pbk. : alk. paper).—
ISBN 0-8018-5483-0 (v. 2. : alk. paper).— ISBN 0-8018-5481-4 (v. 2. ; pbk. : alk. paper),
i. Rome— Religion. 2. Etruscans— Religion. I. Title.

BL802.D813 1996

292.07— dC20

96-14884

A catalog record for this book is available from the British Library.

CONTENTS

VOLUME ONE

Foreword, by Mircea Eliadb

Preface

Abbreviations

xxiii

Preliminary Remarks

1 Uncertainty of the History of the First Centuries 3

2 Political History and Religious History 13

3 The Most Ancient Roman Religion: Numen or Deus? 18

4 Characteristics of the Roman Gods 32

5 The Lost Mythology: The Example of the M atralia 47

6 From Mythology to History 60

7 The Indo-European Heritage at Rome 79

8 Conservatism of Roman Religion : The Case of the luges

auspicium 83

9 The Value of Formulary Evidence 89

10 The Roman Scholars 97

11 The So-Called Pontifical Revolution 102

12 Some Characteristics of Roman Religion 113

13 Balances in Roman Religion 118

14 The Vocabulary of the Sacred 129

15 The Purpose of This Book 134

vii

Vlll CONTENTS

First Part

The Great Gods of the Archaic Triad

1 The Archaic Triad: The Documents

2 Interpretation : The Three Functions

3 Jupiter

4 Mars

5 Quirinus

6 The Archaic Triad: Complements

Second Part
Ancient Theology

1 The Capitoline Triad

Capitoline Jupiter

Juno

Minerva

The Capitoline Triad

2 The Fires of Public Worship

3 Frameworks

The Beginnings
The Times
The Places

4 Man

The Living
The Dead

5 Forces and Elements

Third Function

Second and First Functions

141

148

176

205

246

273

283

291

303

306

3ii

327

333

340

356

363

370

390

VOLUME TWO

Third Part
Extensions and Mutations

1 Personified Abstractions

2 The Gods of the Neighbors

CONTENTS

The Gods of the Enemy

424

The Gods of the Merchants

432

The First Greek Gods

441

An Attempt at Chronology

446

Religion during the Second Punic War

457

The First Literary Evidence

490

Ideological Innovations

497

Thrusts and Resistances

512

The Civil Wars and Religion

526

Fourth Part
The Cult

Sacra publica

553

Sacerdotes

576

Signa and Portenta

594

Notes on the Private Cults

61 1

Appendix. The Religion of the Etruscans

625

Introduction

625

The “Libri” of the Etruscans

633

Libri fulgurates

637

Libri haruspicini

649

Libri rituales

660

The Gods

673

Groupings of Gods

684

The World of the Dead

691

Selected Works on Etruscan Religion

695

Index

697

FOREWORD

Georges Dumezil is perhaps the only foreign author to have his first
book translated into English three years after the publication of a
whole volume devoted to the critical discussion of his theories. In-
deed, when C. Scott Littleton published The New Comparative Myth-
ology: An Anthropological Assessment of the Theories of Georges Dumezil
(Berkeley and Los Angeles, 1966), not a single one of DumeziLs
articles had been translated into English — this for a man who was
at that time the author of some thirty volumes and almost two hun-
dred studies and review articles.

And it so happened that the first book to appear in English, though
one of the most recent and most important, does not belong to the
series dedicated to the comparative study of Indo-European religions
and mythologies, the series that put Georges Dumezil at the center of
a long and passionate methodological debate. From 1940 on, he was,
in fact, becoming well known, first and foremost, as the champion of
a new, structural interpretation of Indo-European religious institu-
tions and mythologies which he saw as reflecting a tripartite ideol-
ogy — the interpretation being elaborated specifically in a book
published in 1958 with just this title ( L’ideologie tripartie des Indo-
liuropeens). While enthusiastically followed by some of the most
distinguished scholars from every area of Indo-European studies,
DumeziLs comparative method was also criticized (and not always
in the most relaxed manner) by other specialists.

The resistance to DumeziLs approach, fortunately now being
overcome in many countries, originated probably for three main
reasons: (1) the fact that the discipline of comparative Indo-European
mythology had been hopelessly discredited by the improvisations of

Xll

FOREWORD

Max Muller and his followers; (2) the tendency, general in the first
quarter of the century, to interpret the spiritual and cultural life of
the protohistorical peoples in the light of what was considered charac-
teristic of the “ primitives thus, the well-articulated mythology,

and especially the implied ideological system, attributed by Dumezil
to the early Indo-Europeans seemed too coherent and too “ profound"
for a protohistorical society; (3) the conviction of the specialists in the
particular Indo-European philologies that it is impossible for a single
scholar to master the entire area of Indo-European studies.

All these reasons were based on as many misunderstandings: (1)
Dum6zil did not use Max Muller's philological (i.e., etymological)
method, but a historical one; he compared historically related
socio- religious phenomena (i.e., the institutions, mythologies, and
theologies of a number of peoples descended from the same ethnic,
linguistic, and cultural matrix), and eventually he proved that the
similarities point to an original system and not to a casual survival of
heterogeneous elements. (2) Modern research has exploded the
evolutionist fallacy of the inability of the “primitive" to think ration-
ally and “systematically"; furthermore, the proto-Indo-European
culture, far from being “primitive," was already enriched through
continuous, though indirect, influences from the higher, urban civili-
zations of the ancient Near East. (3) The “impossibility" of mastering
so many philologies is a false postulate grounded on personal ex-
perience or statistical information, but ultimately irrelevant; the
only convincing argument would have been to prove that Dumezil's
interpretation of, let us say, a Sanskrit, Celtic, or Caucasian text
betrays his inadequate knowledge of the respective language.

In an impressive series of books and monographs which appeared
between 1940 and i960, Georges Dumezil has investigated the Indo-
European tripartite conception of society, namely, its division into
three superposed zones corresponding to three functions : sovereignty,
warrior force, economic prosperity. Each function constitutes the
responsibility of a socio-political category (kings or priests, warriors,
food producers), and is directly related to a specific type of divinity (in
ancient Rome, for example, Jupiter, Mars, Quirinus). The first function
is divided into two complementary parts or aspects, the magical and
juridical sovereignty, illustrated in Vedic India by Varuna and Mitra.
This basic ideological configuration of the proto-Indo-Europeans has

FOREWORD

been differently developed and reinterpreted by the various
Indo-European peoples in the course of their separate histories. For
example, Dumezil has convincingly shown that the Indian mind elab-
orated the original scheme in cosmological terms while the Romans
“historicized” the mythological data, so that the most archaic, and
the only genuine, Roman mythology is to be deciphered in the
“historical” personages and events described by Livy in the first book
of his Histories.

Dumezil has completed his thorough study of the tripartite ideology
in a number of monographs on Indo-European rituals and on Vedic
and Latin goddesses, and, quite recently (1966), he brought out the
present book on Roman religion, the first of his works to appear in
English. The stupendous erudition and the untiring productivity of
Georges Dumezil constitute one of the most fascinating enigmas of
contemporary scholarship. And perhaps the most depressing aspect
of some of the debates centering on his writings has been precisely the
tendency on the part of certain critics to ignore his oeuvre during their
discussions of specific details. For the important, the decisive, element
in the evaluation of Dumezils contributions is his general system of
interpretation, and one has always to keep in mind the ensemble of
his writings on Indo-European religions and mythologies when
criticizing any of the specific applications of his method.

La religion romaine archalque represents a new phase in DumeziTs
production. It is true that he had earlier devoted some books to
various problems in Roman religion, but he always approached the
subject from an Indo-European comparative perspective. (See, for
instance, the four volumes of Jupiter , Mars , Quirinus , 1941-48; Rituels
indo-europiens a Rome , 1954; Dresses latines et mythes vediques , 1956;
etc.). Now, for the first time, archaic Roman religion is presented in
its totality — though, of course, references to other Indo-European
religions are not wholly lacking. This vast, superb work is not a
textbook, nor is it a collection of monographs loosely integrated. The
author insists on the central place to be given what he rightly con-
siders the most important element in the understanding of any
type of religion, that is, its ideas and representations of divinity: in
sum, its theology. It is refreshing and inspiring to read DumeziFs
criticism of the so-called mana-theory, enthusiastically utilized by
1 1 . J. Rose and other scholars to explain the “origins” and structure of

xiv

FOREWORD

Roman religion. One looks forward to the volume in which Dum6zil
will discuss the different contemporary approaches to the study of
Roman religion. But even in the present work the strictures against
Rose, H. Wagenvoort, and Kurt Latte constitute, in themselves, in-
valuable methodological contributions.

Fortunately, more and more the specialists are accepting and con-
veniently utilizing Dumezil’s method and results. In addition to the
importance of his work — and, for the moment, it is the only new and
significant contribution to the understanding of Indo-European
religions — the example of Dumezil is no less important to the disci-
pline of the history of religions. He has shown how to complement a
meticulous philological and historical analysis of the texts with in-
sights gained from sociology and philosophy. He has also shown that
only by deciphering the basic ideological system underlying the social
and religious institutions can a particular divine figure, myth, or
ritual be correctly understood.

Mircea Eliade

The University of Chicago, August 1969

PREFACE

To the colleagues and scholars who follow my work I must say briefly
why, as a mythographer and a comparatist, I am here taking on a task
which is traditionally reserved for Latinists or for archaeologists,
along with the risks which this usurpation involves.

At the moment when the proposal was made to me, seven years
ago, to write the Roman volume for a collection on the religions of
humanity, it proved to correspond to a double need, I might say a
double necessity, of my own research.

Ten years had passed since the publication of my little book
Vhiritage indo-europien d Rome , the ambitious title of which was
certainly premature. During those ten years, I had not ceased to call
in question the results which had been set forth and to approach, in
extended order, a quantity of new comparative problems. The bal-
ance sheet of ff the heritage’’ underwent a transformation. On the one
hand, it was considerably enlarged: the four 1949 chapters had given
to many the impression that, outside of what is covered by the com-
bined names of Jupiter, Mars, and Quirinus, the Indo-European
comparison contributes very little to the exegesis of religious life at
Rome. I believed so myself: between 1938 and 1949, attending to the
most urgent concerns, I had concentrated my investigation on this
central area. But, in the following years, consideration of very diverse
rituals, of several apparently isolated figures in the theology, and of
important religious ideas having no particular connection with the
iripartition showed, on the contrary, the breadth of the material
lying within the scope of the comparison. On the other hand, under
this new illumination, points which had until then seemed to me to
be essential in the very area of the tripartition and on which I had

XVI

PREFACE

extended and reopened discussions lost their usefulness in my eyes:
for example, the question of the meaning, functional or not, of the
three primitive tribes of Rome. As my work proceeded, I gained a
clearer awareness of the possibilities, but also of the limits, of the
comparative method, in particular of what should be its Golden Rule,
namely, that it permits one to explore and clarify structures of thought
but not to reconstruct events, to " fabricate history,” or even pre-
history, a temptation to which the comparatist is no less exposed, and
with the same gloomy prognosis, than the philologist, the archaeolo-
gist, and of course the historian. The proposal which had been offered
to me encouraged me to press forward systematically both with my
research and my revision. Several years of seminars at the Ecole des
Hautes Etudes (Religious Sciences) and of courses at the College de
France were devoted to this work. Two series of reports submitted to
my young comrades on the rue d'Ulm, and the discussions of these
reports which several of them conducted, were particularly profitable
to me: on "Jupiter-Mars-Quirinus,” on the agrarian Mars, on
"flamen-brahman,” and on a great many subjects which have occu-
pied me for almost thirty years, there will be found here, substituted
for the first and second drafts, the clearer, more rigorous, sometimes
entirely new version resulting from this examination.

Since L’heritage, another necessity had become apparent to me,
cutting across the resistance or the reserve which that book had
provoked in a number of favorably disposed Latinists. It is not enough
to extract from early Roman religion the pieces which can be explain-
ed by the religions of other Indo-European peoples. It is not enough
to recognize and to present the ideological and theological structures
which are shown by the interrelations of these blocks of prehistoric
tradition. One must put them back in place, or rather leave them in
situ , in the total picture and observe how they behaved in the different
periods of Roman religion, how they survived, or perished, or became
changed. In other words, one must establish and reestablish the
continuity between the Indo-European "heritage” and the Roman
reality. At a very early stage I had understood that the only means of
obtaining this solidarity, if it can be obtained, was to change one's
viewpoint, to join those whom one had to convince. Without surren-
dering the advantages of the comparative method, or the results of
Indo-European research, but by adding to this new apparatus, in no

PREFACE

XVII

order of preference, the other traditional ways of knowing, one must
consider Rome and its religion in themselves, for themselves, as a
whole. Stated differently, the time had come to write a general history
of the religion of the Roman Republic, after so many others, from the
Roman point of view. The editor s proposal gave substance to this
project, the breadth of which frightened me. In the synthesis presented
in this book, “the Indo-European heritage ” is only one of many
elements, in harmony with the others. Moreover, the service per-
formed by this cohabitation of the new and the old is not one-di-
rectional: if a few excesses of the first comparative inquiries are
curtailed by it, only the recognition of the Indo-European heritage,
carefully delimited, in turn limits the freedom which for half a
century, in France and abroad, has been readily given to archaic
Roman “history,” especially religious history. The survey presented
here is determinedly conservative, justifying a host of ancient facts to
which uncontrolled criticism and the fantasy of the schools and of
individuals had granted themselves all the rights. We are at one of
those reassuring moments which all human sciences experience
more than once in the course of their development, when new
points of view and new tools of observation rediscover the freshness
of the old landscapes, at the expense of the mirages which had been
substituted for them. Along with these mirages, one part of the
difficulty which seemed to separate “Indo-European Rome” from
historical Rome also disappears.

One point is still sensitive and painful, and will remain so for
a long time. The welding of my work with “reality” will be more or
less easy according to one's ideas, based on archaeology, of the pro-
tohistoric and prehistoric periods of Rome. To tell the truth, in this
regard too, the sometimes spirited debates which I have carried on
lor fifteen or twenty years no longer seem so important to me.
In every way, whatever Roman protohistory may have been, and
even if one chooses to include the Sabines in it, the actual events
have been covered, or rather reconstructed, by traditional ideology
,md by the legends which it has produced in the annalistic tradition.
Above all, the real disagreement among the greatest names in
Roman archaeology over the question of origins proves clearly that
the speculations which certain followers of this discipline boldly call
“facts” must still undergo many tests before they merit this great

Xviii PREFACE

name. My personal preferences tend toward the sober and rigorous
method of A. von Gerkan and H. Miiller-Karpe. The latter’s two short
volumes, Vom Anfang Roms (1959) and Zur Stadtwerdung Roms (1962),
the fifth and eighth Ergdn^ungshefte of the Rdmische Mitteilungen ,
seem to me well able to exorcise many demons. Since I have mentioned
welding, it seems to me that the first part of the present book could
be added without difficulty, as a fifth chapter, to Miiller-Karpe’s
1959 book, provided the author would admit that the material traces
cataloged in his chapter 4 — “Menschenfiguren, Beigefasse, Hausur-
nen” — do not yield to us the whole, or the essentials, of the earliest
religion. On the problem of the original Sabine component and the
earliest peopling of the Quirinal, in particular, I feel that I am in
full accord with what is said on pages 38-39 (cf. pp. 44-46 of the 1962
book) :

The old conception of von Duhn, according to which we have, with the
tombs of the Quirinal and the Esquiline, direct evidence of the Sabines, while
the greater number of those in the Forum are evidence of the Latins, is no
longer tenable today. It has been very generally recognized that it is extra-
ordinarily difficult, and even altogether impossible, to establish equivalences
between cultures revealed by archaeology and groups of languages or ethnic
units (. Rdm.-Germ . Forschungen 22, 44). Above all, one should get rid of the
idea that the funeral rites, in the first Italic Iron Age and especially at Rome,
may be considered ethnic criteria. The differences in forms and practices
among the three groups of burials which can be observed at Rome are ob-
viously to be interpreted in terms of chronology, not of races.

The future of these studies would be assured if specialists in the
various disciplines which contribute to the knowledge of early
Rome were willing to take into account the file of problems and solu-
tions which the comparatist here sets before them, in order to specify
or improve it. Unfortunately we are far from this happy collabora-
tion. One is baffled to see a Kurt Latte write a handbook of Roman
religion, or a Carl Koch edit the article “Quirimis” in the Real -
J Encyclopcidie, without deigning to mention the existence of the
Umbrian Jupiter-Mars-Vofionus triad, which prevents explaining the
Roman Jupiter-Mars-Quirinus triad through reasons peculiar to Rome
(see below, pp. 149-50).

Be that as it may, this book, with its sequel, mentioned on pages
137 - 33 , will be my last summing-up in the field of Roman religion:

PREFACE

XIX

in fifteen or twenty years, it will no longer be I who makes the evalu-
ation; I confidently turn over this job to my juniors. Here is how I
envisage the management of this final period of activity. If the labors
of Werner Betz exempt me from making a reevaluation similar to
the present work for the Germanic world, I should like to attempt, in
the Vedic domain, to make the necessary insertion of the comparative
results in the body of the data: I shall probably not have time for
this. More urgent are two books on the epic, the first of which is
to be published by Editions Gallimard, under the title Mythe et
Epopee , vol. i, V ideologic des trois fonctions dans les epopees des pcuples
indo-europeens. A volume on Jupiter-Mars-Quirinus, not definitive
but brought up to date, will follow, and also a book on the theology
of sovereignty (to be published by the University of Chicago Press),
in which my early essays on Mitra-Varuna, Aryaman, and the " minor
sovereigns” will be revised and partially changed. Finally, in the
spirit of liberty and equity defined at the end of “ Preliminary Re-
marks,” I hope to offer for the use of young people a historical account
of these studies, the progress of which has been neither straight nor
easy; also to examine the work of my adversaries, with the aim of
clarifying, and in part of justifying, their opposition, which has
sometimes assumed unusual forms; and more generally to convey
my testimony concerning the masters of my youth and also concern-
ing the scholarly world which I have witnessed or experienced.

Comprehensive surveys of the religion of the royal and republican
periods are very numerous, and several of them (A. Grenier, H. J.
Rose, and others) will be mentioned in the body of the present book.
Besides Georg Wissowa, Religion und Kultns der Rotner, 2d ed., 1912,
and Kurt Latte, Romische Religionsgeschichte , i960, the reader should
consult especially the following works, which expound very diverse
points of view:

Altheim, Franz. Romische Religionsgeschichte . 2d ed. 1956. Vol. 1, Grundlagen

und Grundbegriffe . Vol. 2, Der geschichtliche Ablauf
Bailey, Cyril. Phases in the Religion of Ancient Rome . 1932.

— . “Roman Religion and the Advent of Philosophy/* In The Cambridge

Ancient History 8 (1930) : 423-65.

Bayct, Jean. Histoire politique et psychologique de la religion romaine. 1 957. 2d ed.

1969.

PREFACE

David, Maurice. La religion romaine. 1948.

Fowler, W. Warde. The Religious Experience of the Roman People . 1911.

. The Roman Festivals of the Period of the Republic . 1899.

Grant, Frederick C. Ancient Roman Religion . 1957.

Radke, Gerhard. Die Gdtter Altitaliens. 1965.

Turchi, Nicola. La religione di Roma antica . 1939.

This book was originally intended for a German collection, and the
manuscript — which I continually updated — was sent to the publisher
in 1963. As delays in translation stretched out with no foreseeable
end, I reacquired my rights to the book, and I thank M. Pidoux-
Payot for the diligence with which he immediately undertook its
publication in France. I wish also to express my gratitude to Editions
Gallimard and to the Presses Universitaires de France, who have
authorized me to reproduce in the "Preliminary Remarks” several
passages from earlier books, thus sparing me the hazardous effort
of expressing, in a different way, ideas which have not changed, A
young Japanese scholar, Mr. Atsuhiko Yoshida, has been kind enough
to help me in the preparation of the index, and in this connection
has given me valuable advice, from which I have still been able to
profit.

Georges Dum£zil

Istanbul, September 1966

I am very pleased that the University of Chicago Press decided to
make this work available to the English-speaking public, and for his
encouragement in this respect I wish to thank my friend and colleague
Mircea Eliade. In this translation some passages have been modified
and some discussion added in the notes.

I am happy to have found in Mr. Philip Krapp a competent and
dedicated translator.

Most of the English translations of passages from Latin and Greek
authors are, with a few minor variations, taken from the Loeb
Classical Library translations and are reprinted by permission of the
publishers, Harvard University Press, Cambridge, Massachusetts,
and the Loeb Classical Library from: Augustine, City of God y trans.
W. M. Green; Cicero, De divinatione and De senectute , trans. W. A.
Falconer, De rntura deorum, trans. H. Rackham, De Republica , trans.
C. W. Keyes; Dionysius of Halicarnassus, Roman Antiquities , trans.

PREFACE

XXI

E. Cary; Herodotus, History of the Persian Wars, trans. A* D. Godley;
Horace, Odes and Epodes , trans. C. E. Bennett; Livy, Roman History ,
trans. B. O. Foster, F. G. Moore, and E. T. Sage; Ovid, Fasti , trans.
J. G. Frazer, Metamorphoses , trans. F. J. Miller; Pausanias, Description
of Greece , trans. W. H. S. Jones; Plutarch, Parallel Lives , trans. Berna-
dotte Perrin; Silius Italicus, Punica, trans. J. D. Duff; Suetonius, Lives
of the Caesars , trans. J. C. 'Rolfe; Varro, On the Latin Language , trans.
R. G. Kent; Virgil, Aeneid , trans. H. R. Fairclough; Vitruvius, On
Architecture , trans. Frank Granger.

Volume I of my Mythe et Epopee , mentioned above, was published
in 1968.

G.D.

University of Chicago, February 1970

ABBREVIATIONS

Aelian.

Anim.

V.H.

App.

B.C.

Pun.

Apul. Mag.
Arn. Gent.
Aug. Ciu. D .
Aur. Viet. Or.
Caes. B.G.
Cass. Dio
Cat.

Cato Agr.
Cels.

Censor.

Cic.

Amic.

Arch.

Cad.

Diu.

Ep. ad Brut.
Fin.

Font.

Har. resp.

Greek and Latin Authors

KXavStos AlXiavos (Aelianus)

IJepl i^tpojv ( De natura animalium)

IJoLKiXrj laropla (Varia historia)

'Airmavos ( Appian )

'PwpLa'LKa ipLtpvXia, A-E ( Bella ciuilia )
KapxrjSoviKa ( Bella Punica)

L. Apuleius, De magia

Arnobius, Disputationes aduersus gentes
Aurelius Augustinus (St. Augustine), De ciuitate Dei
[Aurelius Victor], Origogentis Romanae
C. Julius Caesar, De bello Gallico
A Iojv 6 Kcloolos (Dio Cassius), 'Pco/xai'/cT) loropia
C. Valerius Catullus, Carmina

M. Porcius Cato (Cato the Censor), De agricultura
A. Cornelius Celsus, De medicina

Censorinus, De die natali
M. Tullius Cicero
De amicitia
Pro Archiapoeta
Pro Caelio
De diuinatione
Epistolae ad Brutum
Dejinibus
Pro Fonteio

De haruspicum responso
xxiii

XXIV

ABBREVIATIONS

Leg.

De legibus

Leg. agr.

De lege agr aria

Mil.

Pro Milone

Mur .

Pro Murena

Nat . d.

De natura deorum

Philip .

Philippicae

Scaur .

Pro Aem. Scauro

Sen*

De senectute

Sest.

Pro P. Sestio

Vat.

In Vatinium

Claud. B. Get.

Claudius Claudianus (Claudian), De bello Getico

Columella

Lucius Junius Moderatus Columella, De re rustica

Conon Narr.

Kovcjv (Conon Mythographus), Ai^yriaeis ( Narrationes )

Diod. Sic.

A loStupos 6 EiKeXuoTrjs (Diodorus Siculus), 'Iaropicov

pL^XioQrjKrj

Dion.

Aiovvoios 'AXiKapvaacrevs (Dionysius of Halicarnassus),
' PcDpLaiKr} apxcuoXoyta

Fest. L 2 (Paul. L 2 )

S. Pompeius Festus, De uerborum significatu (Pauli
Diaconi epitoma), 2d ed. of W. M. Lindsay = Glos -
saria Latina 4 (1930) : 91-457

Flor.

L. Annaeus Florus, Rerum Romanarum epitome

Gell.

Aulus Gellius, Noctes Atticae

Gloss. ( Lat .)

Glossaria (Latina), iussu Academiae Britannicae edita

Herod.

* HpoSoros (Herodotus), 'IoropLrjs a TToheZis

Herodian.

'Hp<o8uxv6s (Herodian), Trjs pera MapKov fiaoiXeias
ioToplou

Hor.

Q. Horatius Flaccus (Horace)

ad Pis .

ad Pisones

Carm .

Carmina (Odes)

Ep.

Epistolae

Epod.

Epodi

Sertn.

Sermones (Satires)

ps.-Acro in Hor.,

pseudo-Acro, Porphyrio, Scholiae in Horatium

Porphyr. in Hor.

Hyg.

C. Julius Hyginus, Fabulae

Isid. Etym.

Isidorus Hispalensis (Isidore of Seville), Etymologiae (siue
Origines)

ABBREVIATIONS

XXV

Jul. Obs.

Julius Obsequens, Prodigiorum liber

Justin.

M. Justinus (Justin), Pompei Trogi historiarum Philippi -
carum epitoma

Juv.

D. Junius Juvenalis (Juvenal), Saturae

Liv.

Titus Livius (Livy), Ab urbe condita libri qui supersunt

Liv. per ;

Omnium librorum periochae

Luc. Phars .

M. Annaeus Lucanus (Lucan), Belli ciuilis libri ( Pharsalia )

Luer.

T. Lucretius Carus, De rerum natura

Lyd.

* Iojdwrjs Aavpivrios o AvSos (Lydus)

Mens.

Ilepl pLTjv&v ( De Mensibus)

Ost.

Tlepl hiocrripLtiojv ( De Ostentis)

Macr.

Aurelius Macrobius, Saturnalia

Mart. Cap.

Martianus Felix Capella, De nuptiis Philologiae et Mercurii

Non. L.

Nonius Marcellus, De compendiosa doctrina, ed. Lindsay
(Teubner, 1903)

Ov.

P. Ovidius Naso (Ovid)

Am .

Amores

Fasti

Met.

Metamorphoseis

Pont.

Epistolae ex Ponto

Tr.

Tristia

Paul. L 2

See Fest. L 2

Pa us.

FlavoavLas (Pausanias), 'EXAdSos ncpvqy^ats

Pers.

A. Persius Flaccus, Saturae

Petr.

T. Petronius Arbiter, Satyricon

Plaut.

T. Maccius Plautus

Amph .

Amphitruo

Asin.

Asinaria

Aul

Aulularia

Copt.

Captiui

Cas .

Casina

Cure.

Curculio

Men.

Menaechmi

Merc.

Mercator

Mil

Miles gloriosus

Pers.

Persa

Psetul.

Pseudolus

XXVI

ABBREVIATIONS

Rud .

Rudens

Stzc/i.

Stichus

Trin.

Trinummus

True .

Truculentus

Plin. N.H.

C. Plinius Secundus (Pliny the Elder), Naturalis Historia

Plut.

IlXovrapxos (Plutarch)

M or.

'HQlko. (Moralia) :

Defeat, am.

IJepl (f>iAa&eA<f>las (Defraterno amore)

Q.R.

'PayfuuKa (Quaestioties Romanae)

[ Vit .]

plot 7rapaAAr]Aca ( Vitae parallelae) :

Aem. Paul.

Aemilius Paulus

Caes.

Caesar

Cam.

CamiUus

C.Gr.

C. Gracchus

Cor.

Coriolattus

Crass.

Crassus

Mar.

Marius

Marc.

MarceUus

Num.

Numa

Public.

Publicola

Rom.

Romulus

Syll

Sulla

Pol.

IJoMpLOS (Polybius), 'laropiai

Procop. B. Got.

IJpoKomos (Procopius), f Ynep rd>v TroAepaov E-H (De
bello Gothico)

Sen. N.Q.

L. Annaeus Seneca, Naturales quaestioties

Serv. (Serv. II)

Servius (Servius Danielis, interpolator), Commentarii in
Vergilium

SSL It.

C. Silius Italicus, Punica

Solin.

C. Julius Solinus, Polyhistor

Stat. S.

P. Papinius Statius, Siluae

Strab.

Erpdpwv (Strabo), Etoiypa^iKcr.

Suet.

C. Suetonius Tranquillus

Aug.

Vita Diui Augusti

Caes.

Vita Caesaris

Gramm.

De grammaticis

Vesp.

Vita Vespasiani

ABBREVIATIONS

xxvii

Tac.

Ann .

Hist.

Ter. Andr.

Pert.

An.

Idol

Monog.

Nat.

Sped.

Tib.

VaL Max.

Varr.

L.L.

R.R.

Vcrg.

Aen.

Pci

Georg.

Schol Bern., Schol
Veron. in Verg.

Vitr.

AG WG

AIA
\irh.
ML
.1 Nl*

A PAW

ARW

C. Cornelius Tacitus
Annales
Germania
Historiae

P. Terentius Afer (Terence), Andria

Q. Septimius Tertullianus (Tertullian)
Deanima

Deidolatria
De monogamia
Ad nationes
De spectaculis

A. Albius Tibullus, Elegiae

Valerius Maximus, Facta dictaque memorabilia

M. Terentius Varro
De lingua latina
Dererustica

P. Vergilius Maro (Virgil)

Aeneis

Eclogae

Georgica

Scholia Bernensia , Veronensia in Vergilium
M. Vitruvius Pollio, De architectura

Other Abbreviations
Vantiquite classique

Abhandlungender(kdniglichen) GeselbchaftderWissenschaften
Gottingen , Philologisch-historische Klasse

American Journal of Archaeobgy
American Journal of Philology
Archivfiir lateiniscke Lexicographic
Arkivfir nordiskfilologi

Abhandlungen der ( koniglichen ) preussischen Akademie der
Wissenschaften, Philosophisch-historische Klasse

Archivfiir Religionswissenschaft

XXV111

ABBREVIATIONS

Av.

BAB

BCA

BCH

Bomer, Fast.

BPhW

BSL

Catalano

SSR

CIL

Coll. Lat.

AtharvaVeda

Avesta

Acadimie royale de Belgique ; Bulletin de la classe des lettres et
des sciences morales etpolitiques

Bulletino della Commissionearcheologicamunicipale(comunale)
di Roma

Bulletin de correspondance hellenique

Franz Bomer, Ovidius Naso , die Fasten , 2 vols. 1957-58

Berliner philologische Wochenschrift

Btdletin de la Societe de Linguistique de Paris

Perangelo Catalano

Contributi alio studio del diritto augur ale, vol. 1, i960
Linee del sistema sovranna^onale romano , voL 1, 1965

Corpus Inscriptionum Latinarum
Collection Latomus

CPh.

CRAl

Degrassi, ILLR

Dumezil

Aspekte

DIE

Her

Ideol.

JMQ

Classical Philology

Classical Quarterly

Comptes rendus de U Academic des Inscriptions et Belles
Lettres

Charles Darenberg and Edmond Saglio, Dictionnaire
des antiquites grecques et romaines

Attilio Degrassi, Inscriptiones Latinae Liberae Rei Publicae,
vol. 1, 1957, voL 2, 1963

Georges Dumezil

Aspekte der Kriegerfunktion bei den Indogermanen, 1963
(. Aspects de la fonction guerriire che% les Indo-Eurc-
peens , 1956; The Destiny of the Warrior ; 1970)

Les dieux des Indo-Europiens, 1952
Deesses latines et mythes vidiques, 1956
L’ heritage indo-europeen d Rome , 1949
V ideologic tripartie des Indo~Europiens, 1958
Idees romaines, 1969

Jupiter , Mars, Quirinus, 4 vols., 1941-48
Mythe et epopee , vol. 1, 1968
Mitra-Varuna, 2d ed., 1948

QII

RIER

Frazer, Fast .
HOS
HThR
IF

JA
I AOS
IRS
KZ
I at.

Latte

ill

I .ugli, RA
M AAR
MAI
MAL

Mattingly, RC

Mbh.

M/s FR

A In em.

NS
l 'HSR
ritiioi

Plainer, TD
I'M AAR

abbreviations xxix

Naissance d* Archanges, 1945
Naissance de Rome, 1944

Quaestiunculae indo-italicae , 1-16, 1958-61 (in REL and
Lat .)

Ritueh indo-europiens d Rome, 1954
James George Frazer, The Fasti of Ovid , 5 vols., 1919
Harvard Oriental Series
Harvard Theobgical Review
Indogermanische Forschungen
Indo-Iranian Journal
Journal asiatique

Journal of the American Oriental Society
Journal of Roman Studies

( Kuhn's ) Zeitschriftfiir vergleichende Sprachforschung
Latomus

Kurt Latte, Romische Religions geschichte, i960
Lettres dhumanite, Bulletin de V Association Guillaume Budi
Giuseppe Lugli, Roma antica , il Centro Monumentak , 1946
Memoirs of the American Academy in Rome
M imoires de XAcadimie des Inscriptions et Belks Lettres
Monumenti antichi dei Lined

Harold Mattingly, Roman Coins from the Earliest Times to
the Fall of the Western Empire, 1928

Mahdbharata

Melanges d’archeologie et d’histoire de VEcok Frangaise de
Rome

Museum Helveticum

Mnemosyne

Notice degli scavi

Papers of the British School at Rome

Philologus

Samuel B. Platner (and T. Ashby), A Topographical
Dictionary of Ancient Rome , 1927 (= 1965)

Papers and Monographs of the American Academy in Rome

XXX

ABBREVIATIONS

Revue archeologique

RBPhH

Revue beige de philologie et d’histoire

Georg Wissowa, Wilhelm Kroll, et al., Paulys Real-
Encyclopadie der klassischen Altertumswissenschaft

REA

Revue des etudes anciennes

REL

Revue des etudes latines

Rivista difilologia

RhM

Rheinisches Museum fiir Philologie

RHR

Revue de rhistoire des religions

Wilhelm Roscher, Ausfuhrliches Lexikon der griechischen
und romischen Mythologie

Mitteilungen des ( kaiserlichen ) deutschen archaeologischen
Instituts, Romische Abteilung

RPh.

Revue de philologie , de litter ature et d’histoire anciennes

Rg Veda

RVV

Religionsgeschichtliche Versuche und Vorarbeiten

Sat.Brahm.

SatapathaBrahmana

SBBA

Sit^ungsberichte der (] koniglichen ) bayrischen Akademie ,
Philosophisch-historische Klasse

SBHAW

Sit^ungsberichte der heidelberger Akademie der Wissen-
schaften, Philosophisch-historische Klasse

Studi Etruschi

SMSR

Studi e materiali di storia delle religioni

TAPhA

Transactions of the American Philological Association

Tig-

Tabulae Iguuinae

Vetter

Emil Vetter, Handbuch der italischen Dialekte , vol. i, 1953

Vid.

Videvdat

Wiss.

Georg Wissowa, Religion und Kultus der Romer, 2d ed.,
1912

Yt.

Yast

PRELIMINARY REMARKS

UNCERTAINTY OF THE
HISTORY OF THE FIRST
CENTURIES

The history of the religion of the Roman Republic has for a long time
shared in the relative stability which has been recognized in the entire
written tradition of this meticulous people. To be sure, the learned
men of antiquity and, relying on them, a number of Renaissance
scholars, did express doubts, point out contradictions, and emphasize
improbabilities. But this did not touch the heart of the matter: in the
second book of his De republica, Cicero accepted without question the
authenticity of each king of Rome, beginning with Romulus and
Numa, although he was to neglect the fables which ornament the
stories of the reigns. In consequence, European scholars felt some-
what reassured by the hesitations or by the frank declarations of
skepticism with which Livy, Dionysius, and even Plutarch seasoned
the history of the first centuries: was it useful or wise to disregard the
criticism of the men who, with documents at their disposal, had
sensibly evaluated them and, finally, had decided to use them while
noting honestly the limits of their reliability? It was a French Hugue-
not, Louis de Beaufort, tutor to the prince of Hesse-Homburg and
member of the Royal Society of London, who, in a book published
at The Hague in 1738 (second edition, dedicated to the stadholder,

1 75 °). coordinated and amplified the reasons for doubting, not only
with Livy and Dionysius but against them, beyond the frontiers of
their doubt. 1 He sifted the sources which they list, rejected as non-
existent or falsified the Annales Maximi of the pontiffs, the libri lintei,
and the T ables of the Censors, and allowed to stand only the Memorial

1. Dissertation sur I’incertitude des cinq premiers siecles de I’histoire romaine, new ed., with
introduction and notes, by A. Blot (Paris, 1866). The four passages quoted below are re-
spectively on pp. 13, 21, 23, and 179.

PRELIMINARY REMARKS

of the Families, which, however, he challenged as being brazenly
misleading. He emphasized especially the text, indeed worthy of
consideration, with which Livy opened his sixth book:

The history of the Romans from the founding of the City of Rome to the
capture of the same — at first under kings and afterwards under consuls and
dictators, decemvirs and consular tribunes — their foreign wars and their
domestic dissensions, I have set forth in five books, dealing with patters which
are obscure not only by reason of their great antiquity — like far-off objects
which can hardly be descried — but also because in those days there was but
slight and scanty use of writing, the sole trustworthy guardian of the memory
of past events, and because even such records as existed in the commentaries
of the pontiffs and in other public and private documents, nearly all perished
in the conflagration of the City.

In what was reconstituted after the disaster, according to Livy,
Beaufort scents deception :

Among the first decrees [the magistrates] passed was one for searching out
the treaties and laws — to wit, the twelve tables and certain laws of the kings —
so far as they could be discovered. Some of these were made accessible even to
the common people, but such as dealt with sacred rites were kept private by
the pontiffs, chiefly that they might hold the minds of the populace in sub-
jection through religious fear.

The peace treaties? For those from the start of the Republic,
Beaufort has no trouble in setting Polybius against Livy and in
confounding one by means of the other. As for the laws and the
books of the pontiffs, he says, "They did in truth serve to make known
the constitution of the ancient government and to reveal the origin
of certain customs or religious ceremonies; but beyond that they were
of no help in establishing facts, disentangling events and fixing their
dates, which is the essence of history/’

It is remarkable that Beaufort did not push his advantage to its
conclusion; in the second part of his book, in which ‘"the uncertainty
of the principal events of Roman history” is proved, until the torture
of Regulus, he contents himself with stating that “nothing can be
said with certainty about the founder of Rome,” or about the period
of its founding; like Cicero, however, he does not contest the authen-
ticity of Romulus. Concerning the rape of the Sabine women, which
seems improbable to him, he writes:

HISTORY OF THE FIRST CENTURIES

Is it believable that a prince, handsome and adorned with so many good
qualities, as the historians represent Romulus to us, should have been reduced
to the necessity of living in celibacy, if he had not had recourse to violence
in order to have a wife? This is one of those episodes with which the first his-
torians have seen fit to embellish Roman history; and once it was established,
no matter how devoid of verisimilitude it might be, there has been reluctance
to suppress it lest the history lose something thereby.

And similarly concerning the other kings, the " difficulties over the
number of the tribes and over the age of the Tarquins,” the war of
Porsenna, etc.

It was reserved to German criticism of the nineteenth century to
go beyond this well-bred skepticism, embarrassed by its own power
and by the weapons with which it had equipped itself. Following
Berthold Georg Niebuhr, Theodor Mommsen was not content to
attribute to the first historians a preoccupation with “embellishing”
Roman history; he began by appraising the material of these em-
bellishments. In admirable essays he showed that many legends of
die origins, and some of the most important ones, are explained as
romantic projections into the past of events which occurred some
centuries later. Since we have just read Beaufort’s flimsy opinion
concerning Rome’s first war, the rape of the Sabine women, and that
which ensued, I shall recall one of these demonstrations, which will
also be useful in our own analyses. Mommsen’s Tatiuslegende appeared
in 1886 in Hermes and was reprinted in his Gesammelte Schriften (4
| 1906]: 22-35). Here is the resume, as well as the discussion, of this
article which I made in 1944 :

The principal reason for asserting the “ Sabine component” in the origins of
Koine 2 is the legend of the rape of the Sabine women and of the war between
Romulus and Titus Tatius. There is no smoke without fire, it is said: however
.iltered it may be in its details, this legend bears witness at least to an ancient
« « mtact between the two peoples. We shall see. One frees oneself rather quickly
i mm certain remarks by Mommsen which, rather than the theses later formu-
lated by Ettore Pais, continue to threaten to its roots the “Sabine component”

< il the origins of Rome. Mommsen has shown that in this particular case one
may well observe some smoke without fire.

What should be understood by “the Sabines” of whom the legend speaks?
Almost all the versions agree in explicitly giving this appellation its widest

1. This problem will be considered later, pp. 72-78.

PRELIMINARY REMARKS

meaning: namely, that the Sabines are not the inhabitants of the single city
of Cures. If they spotlight Cures as the home of Titus Tatius and as the center
of the coalition which was formed against Rome, and if a current, though
certainly false, etymology connects Quirttes with Cures, “the Sabines” are
nonetheless the federated whole of the Sabine nation (Plut. Rom. 16.3 and
17. 1 ; Dion. 2.36.3-4; cf. Liv. 1.9.9 ; 10.2; 30.6); in short, the Sabines are what will
later be called the nomen sabinum .

This conception of the Sabines, however, involves the legend in gross
contradictions. If it is to be extricated from these, one must ignore them,
as is usually done; but criticism cannot be so accommodating. The synoecism
which ended the war, the union of the two national cells with their institution-
al, religious, and other chromosomes, is conceivable only if the Rome of
Romulus has as its partner a society of the same degree of greatness as itself,
and not an entire federated nation which would overwhelm it. Moreover, the
name of rex Sabinorum given to Tatius (Liv. 1.10.1, etc.) is not meaningful in the
degree to which the Sabines are the nomen sabinum; in primitive Italy it is the
particular urbes which have reges, and the chiefs of federations never bear this
title. Further, the city of Cures itself, which according to the legend was
essentially or totally engulfed by Rome, along with its king, its people, its
riches, and its name ( Quirites ), nevertheless survived and retains its role in
later history . . . Well?

Well, Mommsen has proposed a solution which remains seductive. As often
happens, Rome seems to have made room in its early history for the pre-
figuration of an important episode in the history of the Republic. At the
beginning of this third century which truly laid the foundation for its great-
ness, Rome, together with the already Romanized Latins, effectively allied
itself, after a hard war, with the Samnites (291 b.c.), and, after a brief campaign,
united with all the Sabine peoples; in 290, Rome gave to the latter the rights
of citizenship sine suffragio , in 268 granted them full equality, and shortly
afterward incorporated them into the newly constituted Quirina tribe. Is it
not this union, of a type then new and of great consequence, which ana-
chronistically gave form to the legend of Tatius, in which, despite the contra-
dictions, the Romans saw' the union of two “nationalities”? To be sure, when
Mommsen uses the passage in which Servius ( Aen . 7.709) affirms that the
Sabines, once accepted into Rome, were citizens but without political rights,
dues excepta suffragii latione , perhaps he pushes the analogy too far, for all the
other authors, from Ennius and Varro to Plutarch and Appian, portray the
fusion of the peoples of Romulus and of Tatius as one based on equal rights
(cf. the well-founded criticism of Ettore Pais, Storia critica di Roma 1, 2 [1913]:
423); at least Servius, or rather his unknown source, thus demonstrates that
the connections between the legend of the origins and the diplomatic event

HISTORY OF THE FIRST CENTURIES

of the third century were known to the Romans of the classical era: what does
he do but reproduce exactly the agreement of 290, the first stage of the union?
But there is hardly any need for such precision in the correspondence; the
“myths” which justify events beforehand do not copy them in detail; what
the annalists intended to signify and prefigure here was the total reconciliation
and fusion of two traditionally hostile peoples, the Latins and the Sabines.
The stages of the process were of minor importance; more useful was a
striking abridgment. This is just what the historians give: in the third century
“Rome” is a shortcut for designating the Latin nation, and “the Sabines” are
the federated whole of the Sabine peoples, including Cures; and, by their
treaties, these two partners realized what Livy says of the Romulus-Tatius
agreement (1.13.4): necpacem modo sed ciuitatem unam ex duabus faciunt ; regnum
consociant, imperium omne conferunt Romam — only with the slight alteration that
the legend literally translates the phrase “recipere in ciuitatem” which, in the
event of the third century, was purely abstract and did not imply immigra-
tion. It is easy to verify that this perspective resolves all the contradictions
in the legend which were pointed out earlier.

To these statements Mommsen has added others which are of less interest
because subjective appreciation appears more prominently in them. Moreover
lie gratuitously regards as secondary the connection between the Sabine war
and the institution of the tribes, which is strongly affirmed by the whole
1 radition. He pushes to an extreme the “politisch-atiologisch” character which
he ascribes to the entire “Quasihistorie” of primitive Rome; he even seems to
1 hink that it is not only the matter of the narration, the political meaning of
the account, and the name of the Sabines which date from the third century,
but that the whole affair is a late, literary invention, and that no ancient
t radition, with or without the Sabines, provided the annalists with primary
material. These excesses in his conclusions should not lead one to disregard
what is serious and striking in the principal pieces of the collection. 3

We shall take up the examination of this legend later, at the point
where Mommsen left it, and with other methods, 4 but let us first
note here the dates of the event set back in time by the fabricators of
1 he history of Romulus: 290, 268. After Mommsen a considerable
number of such anachronisms have been recognized, and they are
all located between the second quarter of the fourth century and the
end of the Samnite wars, roughly between 380 and 270. The reign of
A ncus Marcius, and especially his name, doubtless owes much to the

t. NR, pp. 145-48; I am grateful to Editions Gallimard for permission to reproduce this
l< mg quotation, as well as those on pp. 66-73 an d pp. 116-17.

4. Below, pp. 72-78.

PRELIMINARY REMARKS

rise of the gens Marcia in the middle of the fourth century and to the
events of this period. Ancus is said to have founded Ostia and to have
created salt marshes around it (Liv. 1.33.9). The colony of Ostia was
actually founded somewhere around 335 (Carcopino) and it was
near the saltworks of Ostia that Marcius Rutilius, the first plebeian
to be dictator and censor, defeated the Etruscans in 356. Ancus is
said to have installed the conquered inhabitants of Politorium on the
Aventine (ibid. 2). It has been authenticated that the Aventine was
populated in 340. Certain aspects of the “policy” of Servius Tullius
(who is supported by the primores patrum and favors the plebs [Liv.
1.41.6 and 49.2; 46.1]) opposed to that of Tarquin the Old (who relies
on the patres minorum gentium and the equites [ibid. 35.6 and 35.2 and
7]) seem to be composed after the circumstances of the famous
censorship of Appius Claudius (312-308). The Servian organization
which Livy describes in 1.43 is not earlier than the fourth century, and
the estimate of the sums of money mentioned at the time of this
king’s census seems to be based on monetary values at the end of the
fourth century and the beginning of the third. As for the legend of
Romulus himself, the authenticated establishment of a temple to
Jupiter Stator was made in 294, and it is in the theological ensemble
deployed before the battle of Sentinum (295) that Romulus, with his
twin, is attested for the first time in his traditional rank of Quirinus
— and this god receives a temple in 293 on the Quirinal, in place of
the old sacellum with which he had been contented until then.

The ingenuity of an Ettore Pais and of some others has pushed this
quest for anachronisms very far, too far, but the grouping of the great-
er number of the more probable ones between 380 and 270 allows
us to think that it was at this late time that royal history as Ennius
knew it and as we still read it received its definitive form. For the
first centuries of the Republic, the “uncertainty” is just as great, and
it is aggravated by the falsifications of the great families. The war with
Veii, the Gallic drama, and the entire career of Furius Camillus are
known to us only through reworkings which do not allow us to sense
the actual events. Even those of the fourth century are often disturb-
ing, and it is scarcely until the second half of that century that Roman
history begins to display itself, in the rough, with the minimum of
purity required by this great word.

It is another form of involuntary anachronism which complicates

HISTORY OF THE FIRST CENTURIES 9

the study, not so much of events but of customs, of civilization. The
annalists and their heirs the historians, despite some touches of
archaism, do not try to imagine the ancient Romans of whom they
speak otherwise than with the traits of their own contemporaries.
Anticipating the Augustan poets, they speak, in general terms, of the
humbleness of the beginnings; but their Numa, their Ancus, and
Publicola and Servius live, calculate, and create as they would in the
Rome of the Scipios and the Catos. Contrary to all probability, the
armies engaged in the earliest battles are the later legions, except
for the exact number of men; at the height of the struggle in the
Forum, Romulus vows a temple to Jupiter; and, from the beginnings,
Senate and mob are in opposition and trick each other as they will
do until the time of the Empire. What we call the first Romans, the
famous soldier-workers, are at best no more than Catos, made older
by four hundred years, like the peasant paintings in Dalecarlia
which portray scenes from the Gospels in a Skansen setting and with
people in costumes still worn in that part of Sweden on Sundays.

Such being the texts, modem historians accord them more or less
credit, reduce or enlarge the degree of uncertainty, according to the
natural inclination of their own spirits or by virtue of prejudices
(bunded on other considerations, rather than for reasons arising from
the material itself . 5

Do other sources of information compensate for this weakness of
t he annalistic tradition?

There is almost no foreign testimony: the Greeks did not speak of
the Romans until much later, and the first ones to do so at some
length had more imagination and more enthusiasm, but not more
archives or more critical ability than the Romans themselves. The
opening of Plutarch’s Romulus shows well enough what their work v
was able to do — the multiplying of fables, among which the national
1 1 adition soberly made its choice. For the Etruscan period, a unique
,md very important document, the frescoes of the Francois Tomb at
V ulci, conveys simultaneously a stunning confirmation of the existence
of a mercenary soldier named Mastarna, that is, Servius Tullius,

5. See the sensible reflection of P. Fraccaro, "The History of Rome in the Regal Period,"
IKS 47 (1957): 59-65. On the "linen books," see R. M. Ogilvie, "Licinius Macer and libri
hntei" JRS 48 (1958) : 40-46.

IO PRELIMINARY REMARKS

as the head of Rome, and of the Vipina brothers, that is, Vibenna;
and the proof that the Roman portrayal of the event and of the reign
of this same Mastarna differed completely from the Etruscan presen-
tation. The latter, less concerned and composed among a people who
were at the time more literate, is probably closer to reality.

Epigraphy is silent; with rare exceptions, of which only a very
important one directly concerns religion , 6 nothing exists in written
form, either on stone or on tufa, from the first four centuries of Rome,
and it is only in the second century b.c. that the Vollections make a
substantial contribution to our knowledge of Roman civilization and
history.

There remains archaeology, the balance sheet of the methodical
exploration of the site of Rome, which has been pursued for almost a
century, and which in the past twenty-five years has taken on most
promising dimensions. With regard to history, the results of archaeo-
logical investigations are of the first importance. Not in details, to be
sure, but in the overall picture and primarily in chronology, they
restore credit to the principal dates and to the general divisions of
annalistic tradition which mere examination of the texts does not
corroborate. Here briefly is the picture . 7

6. Below, p. 84.

7. I recapitulate the clear summation of Raymond Bloch, Les origines de Rome (Club du
Livre Francis, 1959), PP- 63-100, where the essentials of the bibliography will be found.
On the archaeological data of the Palatine and the Forum, see the works of S. M. Puglisi,
notably “Gli abitatori primitivi del Palatino attraverso le testimonianze archeologiche e
le nuove indagini stratigrafiche sul Germalo," MAL 41 (1951), cols. 1-138; “Nuovi resti
sepolcrali nella valle del Foro Romano/' Bidletino di paletnologia italiana , n.s. 8, no. 4 (1952):
5-17. Einar Gjerstad, after a complete reexamination of the excavations, has been working
for thirteen years on a monumental corpus of the archaeology of the site of Rome: Early
Rome (Acta Instituti Romani Regni Sueciae, 1st ser. vol. 17): vol. 1, Stratigraphical Researches
in the Forum Romanum and along the Sacra Via (1953); vol. 2, The Tombs (1956); vol. 3, Forti-
fications , Domestic Architecture , Sanctuaries, Stratigraphic Excavations (i960); vol. 4, pts. 1 and
2, Synthesis of Archaeological Evidence (1966). The author supports a revolutionary chronology,
still being debated by the archaeologists, which was set forth around 1952 in the article
"Scavi Stratigrafici nel Foro Romano e problemi ad essi relativi/* BCA 73 (1949-50): 3-19.
In his latest book the author displays traces of Bronze Age culture which would extend
the prehistory of the site to about 1500 b.c. Among his opponents I mention A. von Gerkan,
“Zur Friihgeschichte Roms/’ RhM 100 (1957): 82-97; "Das fruhe Rom nach E. Gjerstad/'
RhM 104 (1961): 132-48; reply by E. Gjerstad, " Discussions concerning Early Rome,"
Opuscula Romana 3 (i960). In any case, the use of the legends made by Gjerstad is not
acceptable, " Legends and Facts of Early Roman History," Studier , Kunglinga humanistiska
vetenskapssamfundet i Lund (1960-61), p. 2. See especially H. Muller-Karpe, Vom Anfang
Roms , KM, supplement 5 (1959), and M. Pallottino, who has reviewed and criticized a large
number of recent publications in "Fatti e leggende (modeme) sulla piu antica storia di
Roma," SE 31 (1963): 3 ~ 37 -

HISTORY OF THE FIRST CENTURIES

Despite indications of earlier human presence, it is not until the
middle of the eighth century that several elevated points of the
Palatine were permanently occupied by villages, direct traces of which
can still be seen. The existence of somewhat later establishments on
the Esquiline and on the Quirinal is hypothetical, deduced from the
existence of a rather large necropolis on the former and of five
isolated tombs on the latter. It is only in the seventh century, starting
in 670, and with a disruption due to floods, that the settlement was
extended to the valley of the Forum, which until then had been used
lor burials. Starting in 650, the Quirinal, Viminal, and Capitoline
oiler rich votive deposits, th efauissae, attesting the existence of cult
places, whether constructed or not, which were used for considerable
periods. The principal fauissa of the Quirinal, for example, near the
eighth-century tombs, contains pottery which dates down to about
s8o. From the middle of the sixth century to the beginning of the
lilth, a clear change in archaeological material sends us incontestably
10 Etruria, and testifies that Rome went through the period of Etrus-
can hegemony and wealth which the annalistic tradition describes.
The basement of the Capitoline temple with three cellae attributed
io the Tarquins has vestiges which date from the start of the sixth
u ntury; and some fragments of drains, perhaps of walls, date from
1 lie same period. If the dates which these discoveries impose do not
coincide with the limits which annalistic tradition ascribes to the
Etruscan kingship, the essential facts are nonetheless confirmed;
(onlirmed also, by the disappearance, around 480, of the luxury
1 '-presented by imported Greek pottery, is the return of Rome to its
m riot Latinity in the first quarter of the fifth century.

All this is extremely valuable; it reveals a history which clarifies

• main cultic facts not accounted for by the annalistic tradition.
For example, the annual festival of the Septimontium (Fest. p. 439
I ‘ ) H was celebrated every 11 December; in this festival sacrifices were
nllrred by the dwellers on the three highest points of the Palatine,
those on the three highest points of the Esquiline, and, seventh,
those of Suburra, excluding the people of the Quirinal, the Viminal,

h There is discussion (and has been since antiquity; for example, Lyd. Mens. 4.155)
mo 1 Ik* topographical definition of the seven primitive monies , and even on the name.
1 A. Holland, ~ Septinumtium or Saeptinumtium V* TAPhA 84 (1953): 16-34. See now J.
I *. h it rt," Le Septimontium et la Succusa chez Festus et chez Varron, un probleme d'histoire

• 1 dr 1 opographie romaines/' Bulletin de Vlnstitut Beige de Rome 32 (i960) : 25-73.

PRELIMINARY REMARKS

the Capitol, the Aventine, and, at river level, the Forum. The
federation thus defined agrees with the state of settlement at the
beginning of the seventh century which can be read from the ground
itself, if, as is probable, the Esquiline was occupied shortly after the
Palatine.

If archaeology thus confirms the outlines of royal history, it in-
directly weakens some of its claims. There is general agreement that
until about 575 Rome was considerably less prosperous than the
neighboring Etruscan cities, despite its commercial traffic with Falerii
and Caere, of which there is ceramic evidence, zt\d that it was in-
capable of the conquests and the expansion for which the annalists
give credit to the kings Tullus and Ancus, not to mention Romulus.
The destruction of Alba by the first is a fiction, as the opening of the
port of Ostia by the second is an anachronism. This negative assurance
is also welcome.

But the information provided by archaeology has its natural limi-
tations. The excavations do not permit us even to glimpse the unfold-
ing of events, of which they uncover only the results; moreover, they
do not inform us, whatever one may say, of the most important
element in the study of civilization and religion : the origin and nation-
ality of the men who occupied the montes and the colies, the homo-
geneity or duality of the most ancient settlement. We shall have to
reexamine this point, this option which controls the interpretation of
religious origins. To these we must now return.

POLITICAL HISTORY AND
RELIGIOUS HISTORY

One of the fortunate circumstances in the study of religious facts has
been that Mommsen, the very man who promoted most successfully
the criticism of the legends of the first centuries of Rome, and after
him another great man, his disciple Georg Wissowa, felt that the
uncertainty of political and military history did not automatically
entail that of religious history. A few reflections will quickly make us
realize this relative independence. While the political and military
past, save in the laws and treaties which have resulted from it, is a
product of the recorded or manufactured past, and is without practi-
cal use, religion is always and everywhere an actual and active thing;
its rites are celebrated daily or annually, its concepts and its gods
intervene in the routine of peaceful times as well as in the fever of
limes of crisis. Moreover at Rome, for a long time if not always,
religion gave employment to numerous persons, groups of specialists
who from generation to generation passed on the rules of the cult and
who were supervised by the pontifex. When temples were raised, the
date of consecration and the circumstances of the vow could hardly be
forgotten. Even a terrible blow like the Gallic disaster can hardly have
interfered with such traditions, which were simple ones, and, in their
ritual aspects, were kept alive by practice. Finally, until the great
priesthoods had become strategic positions in the struggles between
classes or factions, religious science, all-important and all-present as it
was, remained autonomous, subject only to its own rules and its
internal logic. As a result it was less exposed to the falsifying ventures
of pride or ambition than was the recital of secular events. Briefly, to
limit ourselves to a few striking examples, the history of the Tarquins
may be only a tissue of fables, but the authenticity of the Capitoline

PRELIMINARY REMARKS

cult which they bequeathed to the Republic is not thereby compro-
mised. It was certainly not Numa who created the offices of the
great flamens, and especially not the first, the flamen of Jupiter; this
does not prevent the motive which the annalists give for this creation
— to relieve the rex of the greater part of the religious duties which
were incompatible with his necessary freedom of action — from
expressing an affinity between the rex and the flamen Dialis and,
beyond the priest, between the rex and Jupiter. Such an affinity is
entirely consistent with what we know of the actual status of the
flamen.

On the other hand, apart from all theory, Mommsen and Wissowa
understood that a religion would not be essentially, at any given
moment, an anarchic accumulation of conceptions and prescriptions
brought together by accident. To be sure, from one end of its history to
the other, Rome shows a remarkable aptitude for absorbing whatever
religiously powerful gods and festivals its circumstances and surroun-
dings offered it; but whatever it takes thus, it annexes, with pre-
cautions, to a preexisting and already rich national patrimony, in
which natural divisions and partial structures are easily seen, if not a
unitary plan. Wissowa's manual has been criticized for its systematic
arrangement. But this plan results from the nature of the material.
It matters little that this author has labeled the two principal divisions
of his study — ancient gods and imported gods — with the names
Indigetes and Nouensiles. These names have been misinterpreted as
meaning "indigenous” and "newly installed,” an error which is
displeasing to the eye because it is repeated in the running heads of
222 odd-numbered pages; however, one quickly realizes that it has
practically no importance and does not prevent the division from
being useful.

Finally Mommsen and Wissowa were imperturbably indifferent to
the ephemeral theories which succeeded one another during their
lives regarding the nature, the origin, and the stereotyped evolution
of religious facts : the solar mythology of Max Muller or the animism
of Tylor, the spirits of vegetation of Mannhardt and his disciples, or
Salomon Reinach's totemism — all aroused the defiance of these
exact and precise scholars. Wissowa’s resistance was doubtless not
without hubris , and one must regret the summary execution which
he performed, at the end of a footnote, on the author of The Golden

POLITICAL AND RELIGIOUS HISTORY 15

Bough . 1 But this excessive reservedness or, if you will, obstinacy, is
worth more than the fads, both before and after Religion und Kultus
der Romer , which have produced so many studies that are outmoded as
soon as they are written. Wissowa’s manual needs to be brought up
to date and, with regard to its doctrine, corrected in large part.
Nevertheless it remains the best; it has not been replaced.

Those who have applied to the interpretation of Roman religion
the successively or simultaneously fashionable theories of the last
century have all posed, explicitly or not, an identical postulate:
namely, that the earliest Romans, concerning whom tradition says
absolutely nothing valid, were “ primitives/' comparable on an
intellectual level to the peoples of America, Africa, and Melanesia,
who have been observed for only two or three hundred years, most
of them for only one hundred, and who belong not to the historians
but to the ethnographers. Their religion, then, should be cast in the
molds which each school regards as primitive; and it is from these
elementary forms, by the process of evolution, that the religion
professed by the grandchildren of these “ primitives/' the Romans
whom we know through the classical authors, should have emerged.

It must be recognized that the innovators have found accomplices
to their theories in the classical authors. It is a commonplace of the
Augustan epoch to contrast the luxury, the complexity of life during
the great reign, with the simplicity, the truly elemental, almost
embryonic character of primitive Rome, not only in habitat and cus-
toms but in political institutions and cults. If one is to believe these
authors, Rome seven centuries before Augustus was nothing but a
lew hundred shepherds on the Palatine. It will also be recognized that
archaeology reinforces this impression. Seeing the few traces of huts,
tiny and irregular, that are shown to him, what visitor to the Palatine
does not repeat with Propertius (4.1. 9-10):

Qua 2 gradibus domus ista Remi se sustulit olim
Unus eratfratrum maxima regna focus . . .?

Who does not wonder how one of these casae could have sheltered
iwo brothers and a hearth at the same time? In such miserable

1. Wiss., p. 248, n. 3.

2. Var. quod, quo.

1 6

PRELIMINARY REMARKS

dwellings it is hard to imagine a life which was not entirely occupied
by the most pressing needs, in which there would have been enough
freedom of thought to conceive, organize, and preserve a theology
above the “ primitive ” level, however one imagines it.

Nevertheless we must resist this temptation. To judge from what
remains of Emain Macha in Ireland, or from the twenty circular
huts which stand at the foot of Caer y Twr, near Holyhead in Angle-
sey, and from many other sites, it is hard to imagine that there
existed in these Celtic countries a body of druids whose studies —
theology, ritual, law, and epic traditions — lasted up to twenty years.
Over how many archaeological excavations does not one experience
the same astonishment! The Latins who settled on the Palatine were
not, any more than their Celtic cousins, new men with everything
to create and discover for themselves. They were, and their language
proves it, the descendants of invaders from distant parts, coming by
stages, and we cannot today regard them as primitives. Here for the
first time we must write a name which will recur frequently in this
book : they were Indo-Europeans who used the same word for politico-
or magico-religious concepts as the Vedic Indians or the oldest Irani-
ans at the other end of the Indo-European sphere, sometimes the
same word as the Celts used for the same or closely related concepts.
Not to speak of religion, which will be our subject matter, it is re-
markable that the head of the primitive Roman society bears the old
Indo-European title *reg~, like that of the Vedic society ( raj[an ]),
and like that of all the ancient Celtic societies (rig-). This single fact
proves that the dwellers in the cabins on the Tiberine mantes were
not groups of inorganic families who were to be associated, at the end
of a certain period, in creating new institutions, but rather that they
had arrived with a suprafamilial structure and a traditional political
organization. How indeed can one suppose that these men who had
inherited from a distant past the idea as well as the word rex could
have allowed it to fall into disuse and then have reactivated it, under
the same name?

If then we do not directly know the Latin r£x, comparison with the
Irish ri and the Vedic rdj(an) allows us to imagine with some clarity
what the Indo-European *reg- was, from which they are derived. The
function of the rdj(an) is not, as was formerly proposed, “derived”
from that of the head of the family. Like the ri y he transcends social

POLITICAL AND RELIGIOUS HISTORY

divisions, represents powers, and has duties and rights which cannot
pass as mere amplification of the status of the head of a family.
Like the ri with his personal druid, he lives and functions symbioti-
cally with an eminent representative of the priestly class, the brahman
who is his chaplain, hi spurdhita, and who gives back to him in mystical
protection what he gives out of generosity. Not only is his consecra-
tion an important religious ceremony, but he has at his disposal an
entire range of appropriately royal ceremonies, notably the sacrifice
of the horse which makes him a kind of super-king— just as Ireland
recognizes an ardri at the head of its hierarchy of rig . The Roman rex
must not have had a very different relationship to the social body over
which he presided. If what we know of his shadowy descendant in the
time of the Republic, the rex sacrorutn , is not enough to give the
measure of this difference, at least the solidarity of the rex with the
most highly placed of the flamens (Liv. 1.20.2) recalls that of the
rdj(an) or of the ri with their principal brahman or druid ; 3 besides,
republican Rome preserved a meaningful connection, which will be
studied later, between the annual sacrifice of the horse and, lacking
the king himself, the “ house of the king,” the Regia, in a ritual whose
symbolism is close to that of the Indian asvamedha , 4 The exact corre-
spondence of personages who bear the same title, and whose wives
likewise have as their titles archaic derivatives of this word (with a
suffix n: Latin regina , Vedic rdjni-, Irish rigairi), cannot be the result of
chance. There has been a continuity from an Indo-European structure
to the Roman structure which we know, and the societies of the
Palatine, the Esquilfiie, etc., displayed the traditional complex or-
ganization, with a rex duplicated religiously by the first fl amen, and
t raditional royal ceremonies, to whose existence the final placement
< >f the tail and head of the Equus October in a Regia transplanted to the
l ; orum still bears witness.

I confine myself here to these considerations. They suffice to suggest,
not the inaccuracy, but rather the incompleteness of the idea held by
frazer of the Latin kingship, in the line of the Wald - und Feldkulte ,

1 hat the essential element of its office was magical control of the
f ecundity of nature.

1. See below, pp. 151-53 and 580-82.
4 . See below, pp. 226-27.

THE MOST ANCIENT
ROMAN RELIGION:
NUMEN OR DELIS'?

Leaving aside the fantasies based on totemism, which nobody defends
any more, as well as the doctrines of Mannhardt, which we should
endeavor not so much to refute as to integrate, in their limited place,
in a more general and better-balanced view of Roman religion, we
must halt before the only one of these theories which still, in various
forms, holds the spotlight in many books. This is the theory which
has been called, by an artificial and negative word, predeism, 1
or, more recently, by an expressive and more suitable term, dyna-
mism. In the origin of religious representations, it stresses a belief
in a power diffused or rather dispersed throughout numerous ma-
terial supports, and acting at once roughly and automatically, a
belief that should have been older than the notions of personal gods
and even spirits, considered as products evolved from it.

The observation of the Melanesians furnished this doctrine with
its point of departure. In 1891, in his book The Melanesians (p. 118
as quoted in H. J. Rose, Ancient Roman Religion , p. 13), Bishop
Codrington gave a definition of the mana which has since flourished :

The Melanesian mind is entirely possessed by the belief in a supernatural
power or influence, called almost universally mana. This is what works to
effect everything which is beyond the ordinary power of men, outside of the
common processes of nature; it is present in the atmosphere of life, attaches

1. K. Vahlert, Praedeismus und romische Religion , Diss. Frankfurt (1935). In this form,
the theory has hardly any supporters. Specialists in the history of religion may regard the
following discussion (pp. 18-46) as useless: they have long known that “ primitive” thought
is not such. But the assertions of H. J. Rose, A. Grenier, and others continue to impress
many classicists who are less aware of the results of modern ethnography.

“numen” or “deus”? 19

itself to persons and things, and is manifested by results which can only be
ascribed to its operation.

In 1926 H. J. Rose, an English Latinist, was enlightened by this
revelation of the most elemental form of the sacred. He quickly
amassed materials by which to prove that this same conception
dominates Roman religion and, taken back to the origins, explains
its entire development. 2 He sought the name which the Romans gave
to the mana and found it : it is numen. Rose’s influence was considerable
in the ensuing quarter of a century.* At Utrecht he received strong
support in the person of H. Wagenvoort, whose book Imperium:
Studien over het mana-begrip in %ede en tool der Romeinen (1941) 4 Rose
translated under the title Roman Dynamism (19 47). In France, A.
C.renier gave Rose enthusiastic approval. 5 Some prominent Hellen-
ists, won over by the contagion, have undertaken to identify the mana
in their domain, and have found it under a name just as unexpected
as numen. It is, they say, Salfxwv.

Various indications lead us to believe that the height of the craze has
passed and that the theory will not long survive its promoter, who
died in 1961. It still holds strong positions, however, and must be
examined with care. Here is how Rose stated it in 1926 ( Primitive
Culture in Italy , pp. 44 - 45 ). after defining classical Roman theology,
‘except for one or two great gods,” as a “polydaemonism,” that is,
an accumulation of beings, each capable of accomplishing a single
action but without any existence, beyond their specialty, either in
1 he cult or in the imagination :

. . . They are not so much gods as particular manifestations of mana . Spinen-
sis provided the mana necessary to get thorn-bushes (spinae) out of people’s
liolds; Cinxia, that needed for the proper girding (cingere) of the bride; and
so with innumerable others. What stories could anyone tell about such
phantasmal, uninteresting beings as these?

2. Before Rose, Frederic Pfister, BPhW 40 (1920) : 648.

3* I have had occasion several times to debate, not very pleasantly, with this author,
notably RHR 133 (1947-48) : 241-43; DL, pp. 41 and n. 2, 118-23.

4. I have examined chapter 4 of this book in 4 ‘ Maiestas et grauitas, de quelques differences
< inre les Romains et les Austronesiens,” RPh 26 (1952): 7-28; (see now JR, pt. 1, chap. 6)
whence an exchange: H. Wagenvoort, " Grauitas et maiestas M nem., 4th ser., 5 (1952):
>87-306; G. Dumezil, “ Maiestas et grauitas, H,” RPh. 28 (1954): 19-20; cf. O. Hiltbrunner,
"Vir grauis,” Festschrift A. Debrunner (1954), p p. 195-206. I shall shortly examine two
<>1 tier chapters of Roman Dynamism.

5. “Observations sur l’un des elements primordiaux de la religion romaine,” hat . 6
(1947): 297-308.

PRELIMINARY REMARKS

The Latin word for them, or for their power, is interesting; it is numen.
The literal meaning is simply “a nod/’ or more accurately, for it is a passive
formation, “that which is produced by nodding,” just as flamen is “that which
is produced by blowing,” i.e., a gust of wind. It came to mean “the product or
expression of power” — not, be it noted, power itself. Properly speaking, the
gods, and sometimes other powers more than human, or than ordinary
humanity, have numen; but as their business is just to have numen and nothing
more, they are themselves often called by that name, especially in the plural,
numina. As the theological thought of Rome advances, the work tal^es on a
higher meaning, and comes to signify “divinity,” “deity”; but we are not
now concerned with this part of its development.

But it might well be said that to conceive of a spirit of any sort, even if he
does nothing more exalted than to give farmers, from time to time, power to
perform successfully the important business of manuring their fields (over
which the god Stercutius presided), represents a not inconsiderable effort of
abstract thought. Indeed, priestly theologians in Rome laid hold of these
ancient numina and extended and classified them, adding many of their own
invention, until they might almost be said to form a list of the detailed func-
tions of Deity in general, or of the various ways in which his help might be
sought. But their history can be traced back to a much lower stage than that,
indeed to nothing higher than the savage concept of mana locally resident in
some place, or in some material object.

Later, in Ancient Roman Religion , published in 1950 (pp. 21-22),
Rose claimed that he could specify how the ancient Romans, starting
with this notion of numen-mana, arrived at the conception of personal
gods:

Since numen is found in sundry places and attached to various persons and
things, it is not remarkable that its manifestations were sometimes less and
sometimes more potent. If they were strong, and especially if they were
regular in their occurrence, the natural conclusion would be that they were
produced by a kind of person who had much numen and was ready to display
it for the benefit of those who approached him in the proper manner. This
person was a god or goddess, and concerning the nature of these beings the
Roman, left to his own devices, seems to have had a great incuriosity.

Before examining this thesis and the principal facts which have been
advanced to support it, it is proper to recall two fundamental con-
cepts in the study of religions.

The first is that in all religions, including the most highly developed,
the faithful practitioners do not all practice their faith on the same

“numen” or "deus”?

level; on the contrary, a wide range of "interpretations” can be
shown in a single epoch or in a single society, often extending from
pure automatism to the subtlest mysticism. Mircea Eliade has made
an excellent formulation of this observation:

The manipulation of the sacred is, by itself, an ambivalent operation,
particularly in this sense, that the sacred may be tested or experienced,
whether on the religious level or on the magical level, without the manip-
ulator's necessarily being clearly aware of what he is performing — an act of
worship or a magical performance. The coexistence of the magical and of the
religious within a single consciousness is moreover extremely frequent. The
same Australian who knows that there is a supreme being (Daramulun,
Baiame, or Mungangao) living in the sky, whom he invokes during initiation
ceremonies, practices magic assiduously, a magic in which none of the gods
is present. A present-day Christian prays to Jesus Christ and to the Virgin
Mary, which does not prevent him from occasionally manipulating the
sacred images (for example, if there is a severe drought) for purely magical
purposes. Thus the immersion of sacred statues is supposed to bring rain,
etc . 6

Still more: the same individuals and the best-educated, according
to circumstances, "practice” on various levels, sometimes with the
insights of theology, sometimes with a simple, voluntarily childlike
trust in the efficacy of their actions or their words. Thus every religion,
of whatever kind, continues to produce, as long as it exists, attitudes
and behavior which may equal or surpass in simplicity those of the
primitives, not merely in appearance but in reality. One must,
however, guard against seeing in such attitudes and behavior a
survival, much less the origin or the remains of a pattern from which
everything else has evolved. That which is typologically primitive may
not be chronologically so. Doubtless it would be better to speak of
"rough” or "inferior” forms, using these terms, however, without a
pejorative connotation.

The second concept, bound up with the preceding one, which the
student should bear in mind is that the concrete apparatus by which
religions materialize the invisible or enter into communication with it
is limited. To live his faith, man makes use only of his sensory organs
and limbs, and to portray the object of that faith he uses only the
resources of his industry. We shall return shortly to the portrayals,

6. Les techniques du Yoga (1948), p. 227.

PRELIMINARY REMARKS

but shall consider here only the actions. The godfather and godmother
at a baptism touch their godchild just as the Roman magistrate
touched the doorpost of the temple which he was inaugurating. The
communicant consumes the host in the same way as the faithful in a
great number of cults on every level consume a sacred food, animal
or vegetable. The confirming bishop slaps the candidate for con-
firmation just as the ancient Roman, in the ceremony of manu-
mission known as festuca, gave a slap to the person whom fie was
manumitting, and just as the brahmans, in the Indian ceremony
of royal consecration, struck the king. And so on. This does not mean
that here and there the level, the intention, the very motive of the
ceremony are the same. There is a considerable difference between
the ideological substratum of the eucharistic communion as prac-
ticed by the Catholic who believes in the actual presence and as
practiced by the Protestant who sees in the Lord’s Supper nothing but
a commemoration; there is an even greater difference between what
both expect as a spiritual profit from this act and the physical and
racial strengthening which a totemic food is claimed to give to who-
ever consumes it. I doubt that the godparents imagine they are
communicating a fluid to the baby whom they touch; their action
simply shows that they are speaking and pledging themselves in its
name — it is not efficacious, but symbolic. Wagenvoort wrote an
entire and very interesting chapter on contactus, the ritual actions
involving touch, in Roman religion. But if it is possible that the touch
of the consecrator of a temple ( postern tenere) transmits a force, I
doubt that this would be true of the touch of the general who per-
forms the deuotio, during which, among other elements of a complex
setting (a veil over his head, his hand on his chin), he must stand on a
spear. One should thus be circumspect in one’s interpretation of a
ritual action concerning which there is no explicit and reliable textual
commentary. Unfortunately, this is usually the case with Roman
rites.

Having said this, we cannot deny that many of the features of our
Roman documentation conspire to give the religion a “primitive”
appearance. Some relate to particular gods, others to the general
characteristics of the gods. The former, to which the school of H. J.
Rose assigns the greatest importance, and which concern Mars and
Jupiter, are not, however, the most striking.

"numbn” or “deus”?

If one goes along with Rose and his followers, Mars was at first
only a lance, and Jupiter, at least under one of his principal aspects,
a stone, each object carrying in itself an appreciable supply of mana.
We shall confine ourselves to a discussion of Mars and his lance, or
lances. There are two kinds of facts here. 7

On the one hand, at Rome, as in other Latin cities, one or more
hasta(e) Martis were preserved, which sometimes moved by them-
selves, apparently making a noise and thus announcing dangerous
happenings. Following the official announcements of prodigies the
historians often comment: hastae Martis in Regia sponte sua motae
sunt .

What do the words sponte sua mean? The primitivists interpret
them strictly: it is by their own power, by their mana , without divine
intervention that the lances shake, thus preserving in historical times
a predeistic representation, a witness of the time when a personal god
had not yet freed himself from the mana of the weapon. However
one may understand sponte sua , this interpretation is not legitimate.
It is in fact possible that the lances shake “by themselves/’ in the
literal sense of the word. But this belief would recall an order of
representations known among religions which are far removed
from primitivism. In Scandinavia, for example, at the end of the pagan
era, the Njdlssaga (30.21) speaks of the halberd of Hallgrimr which
made a loud sound whenever a man was going to be killed by it,
“so great was the power of magic — ndttura [ Zauberkraft , designated
by the Latin word natural]— which there was in it/* Like this halberd,
ihe hastae would simply be magical objects, and this banal statement
would not inform us about the evolution of theology or the origin of
Mars any more than the spontaneous noise made by the halberd
e xplains the origin of the gods of battle, 6&inn and )>6rr. Since,
however, the lances are always presented as “the lances of Mars,”
a personal god of whom it is hard to deny that he has some connec-
1 ion with battle, another interpretation recommends itself. The words
sponte sua in the catalogs of prodigies are not intended to deny the
intervention of the invisible god but only that of men and of any
palpable agency or motive power; thus they would mean “without

7. Here I develop and improve a discussion from my DIE , pp. m-17. A good criticism
*»l numen-mam by S. Weinstock appeared in JRS 39 (1949): 166-67; another, by P. Boyance,
in the Journal des Savants , 1948, pp. 69-78, and in Vinformation littiraire, May-June 1955,
pp. 100-107.

PRELIMINARY REMARKS

having been touched or moved by any person/' There is no lack of
parallels for this use of the expression. In the fifteenth book of the
Metamorphoses Ovid recounts the birth of Tages, the tiny mannikin who
is said to have founded Etruscan divination and who emerged one
day from a clod of earth under the astonished eyes of a peasant at work
in his fields (lines 553-59). According to the poet, the poor rustic
"saw in his fields a clod, big with fate [fatalem glaebam ], first moving
of its own accord, and with no one touching it, then taking dn the
form of man and losing its earthy shape, and finally opening its new-
made mouth to speak things that were to be.”

sponte sua primum nulloque agitante moueri . . .

Tages is born in the manner of mice, who were thought to be
produced by the earth. But in the first stage of the operation, before
the clod is, so to say, delivered of his child, the laborer sees only one
thing: the unexplained movement of this clod, ttullo agitante , un-
touched by anyone human, anyone of our world. Sponte sua merely
duplicates nullo agitante and apparently does not preclude what the
second stage of the process shows, namely, that someone on the inside,
a little supernatural being, is moving the clod. The same may be
true of the “spontaneous ” movement of the lances, with the unseen
Mars setting them in motion. Many facts recommend a preference
for this interpretation, notably the following. Livy (22.1.11) relates
that at Falerii, shortly before the disaster of Lake Trasimene, sortes
sua sponte attenuatas unamque excidisse ita scriptam: t( Mauors telum
suum concutit .” Plainly what we have here is a warning, a threat
equivalent to those received at Rome or at Lanuvium (Liv. 21.62.4:
the lance of Juno), or at Praeneste (Liv. 24.10.10), when the lance or
lances move; only, at Falerii, it is the divination tablets which move
perceptibly and "by themselves,” and the shaking of the weapon by
the invisible Mars is only declared, not observed. By way of compen-
sation, however, the agent of this movement is also declared: it is
the god himself. Thus at Rome, where only the god is invisible and the
lances are seen, it is still he who moves them when they move "them-
selves.” To sum up, the movement of the lances either suggests the
magic which in all places and times is juxtaposed with religion, or
else it is to be explained, in terms of "deism,” by the action of Mars.
In neither case does anything suggest that this is a survival of "pre-

2.5

“NUMEN” OR <f DEUs”?

deism/’ a vestige of a conception antedating Mars and from which
Mars might have issued.

The second fact connected with the lance or lances of Mars, from
which the primitivists draw important conclusions, is the ritual pro-
cedure performed by the Roman general before he entered on a
campaign (Seryius Aen . 8.3). He betook himself to the chapel in the
Regia consecrated to this god, shook first the sacred bucklers which
hung there, and then the lance of the statue itself, hastam simulacri
ipsius y saying, “Mars, uigilaV * It is very probable, as has been re-
marked, that the setting of the rite is given by Servius in a relatively
recent form : in ancient times Mars no more had a statue than did the
other gods. This then must have been not a lance held in the hand of a
simulacrum which the general touched, but a separate, detached lance,
sufficient to itself Two alternative conclusions by the primitivists:
in the oldest times Mars himself did not serve as mediator, and the
entire transaction was between the general and a lance charged with
matia; or the lance was itself Mars and had, by the process of evolu-
lion, produced a vague personal god who was still partially embodied
in it. They have supported their theory with a text by Plutarch (Rom.
29.2) which says, in a hasty and poorly constructed sentence, that there
was in the Regia. a lance “which was called Mars ” (iv 8e rij Sopv

Kadihpvfievov "Apea Trpoaayop€V€iv)y and with a passage of Amobius
(6.1 1) which follows Varro in noting that the earliest Romans had
pro Marte hastam. On the basis of these brief indications and of the
general’s action, a prehistory of Mars has been constructed in three
phases: first, an inanimate lance was charged with mana , as a storage
battery is charged with electricity; then the importance of this mana
led to the conclusion that a spirit inhabited the lance; and finally this
spirit was separated from the lance and, having become the prime
factor, having become a god, received the lance as a weapon.

Such an interpretation puts a heavy strain on the texts. It is enough
to read in its entirety the passage from Arnobius and also the parallel
passage from the Logos Protreptikos of Clement of Alexandria (4.46)
lo understand what Varro meant: not that the ancient Romans
1 bought the lance to be Mars, but that it replaced a better representa-
tion of Mars. Varro, says Clement, declares that the cultic image of
Mars (rov*Ap€ <09 to £6avov) was formerly a lance “because the artists

2 6

PRELIMINARY REMARKS

had not yet entered upon the track (dangerous in the opinion of the
Christian scholar) which leads to the portrayal of the gods with beauti-
ful faces/' It is clearly in the light of this text that we must understand
pro in the text in which Arnobius lists a number of peoples for whom
an object “took the place ' 5 of a god: lignum Icarios pro Diana , Pes-
sinuntios silicem pro Deum matre , pro Marte Romanos hastam , Varronis
ut indicant M usae atque ut Aethlius memorat ; besides Arpobius im-
mediately repeats the reason, which, like Clement, he /took from
Varro: ante usum disciplinamque fictorum pluteum Samios pro Junone —
the Samians were satisfied, so long as they had not acquired the
custom and the technique of molding clay, with a board “in the guise
of” Juno, in place of and instead of Juno . 8

Some day it will be necessary to restore to the history of religions the
idea of the symbol which is today so underrated and yet of such capital
importance. Symbolization is the basic resource of every system of
thought, every articulate or gestural language. It is what permits one,
if not to voice, at least to approximate, to delimit the nature of things,
by substituting for the stiff and clumsy copula of identity, “to be,”
more flexible affinities: “to resemble / 5 “to have as attribute or
principal instrument,” “ to recall by an important association of ideas.”
Consider the cross in many oratories, the rudimentary crucifix on
which the Christ is not shown but which is merely two pieces of wood
placed at right angles. If someone from the outside, who does not
know of Christianity or wishes to make fun of it, should see this,
how would he judge the often burning devotion of which this cross is
the object? As a variety of “dendrolatry 55 : 9 these pieces of wood,
it might be said, are holy, they emit mana y etc. Of course we well
know that it is an entirely different matter: the simplest cross con-
jures up the passion, the scheme of salvation, from the Incarnation to
the Redemption, Adam with the tree of sin and Jesus with the tree of
forgiveness. The fervor of the suppliant is not directed to the material
object but to the historical realities and to the dogmas which the
Gospels and theology associate with the agony of Calvary. The pieces of
wood are only an aid, a means of recall, precious and even holy to the

8 . Cf. Justin. 43.3.3: mm et ab origine rerum pro signis immortalibus ueteres hastas coluere ,
ob cuius religionis memoriam adhuc deorum simulacris hastae adduntur.

9. I borrow this word from one of the worst descriptions ever made of a religion, based
on a purely archaeological dossier: G. Glotz, La civilisation egecnne (1923), pp. 263-95.

"numen” or "deus”?

degree that what it “ recalls " is precious and holy. Just as the Samians,
before the development of the arts, had a board, pluteum, pro Junone ,
so at Sparta the most ancient portrayal of the Dioscuri was two boards,
<Wava; despite an opinion which is occasionally voiced, one will not
see in this fetish the origin of the divine twins. The wide extent of the
cult of the twins, of which the Spartan cult is only a particular case,
renders this interpretation very improbable. On the contrary, it will
be readily admitted that by understanding them as what they repre-
sent everywhere, that is, two young people, two brothers of the
human species, an already abstract idea or a still rudimentary tech-
nique has expressed the essence of the concept — the fact of the pair
—by the juxtaposition of two parallel bars. 10

Why did not the first Romans portray Mars, or indeed the other
gods? 11 According to Varro, as we have seen, it was only for lack of
artistic maturity. It may also be, as among the Germans whom Taci-
tus described, out of reverence, out of the feeling that any portrayal,
and likewise any "putting into the temple,” would be an imprison-
ment. Finally it may be that they did not feel the need for such por-
trayal. It is not necessary to postulate, as required and reciprocal
equivalences, "personality = outline,” "anthropomorphism = por-
trayal/' Many peoples conceive of gods in human form, and even
know that such a god has such a peculiarity, such a physical mon-
strosity, a red beard, three eyes, an enormous arm, yet do not show
this knowledge in wooden, clay, or stone images. But as it may be
useful to indicate distinctively the presence of this invisible being,
for sacrifices and in sacred places, without any possible confusion with
others, an object is used to identify and characterize it. In Rome this
ob ject may be the hearth fire of Vesta, the flint or thunderbolt of
Jupiter the Thunderer, or the lance of the warrior Mars. Thus, under
.it t ack from all sides, the sentence from the Life of Romulus, stating that
i here was in the Regia a lance "called Mars,” slips out of the hands of
i he champions of the mana theory. Do not these words simply allude
to the rite in which the commander in chief, is qui belli susceperat
atram , shook the lance while saying, "Mars, uigila ,y 7 and is not this in-
junction addressed to the god beyond the symbol rather than to the

i o. See the important study by K. Schlosser, Der Sigttalismus in der Kunst der Naturvdlker
( iosi).

1 1 . There are perhaps reservations to be made concerning the aniconism of the earliest
Komans : P. Boyanc£, in REA 57 (1955) : 66-67, with the bibliography.

PRELIMINARY REMARKS

symbol itself? Confident of being understood, not anticipating H. J.
Rose, Plutarch wrote hastily, without troubling himself over style or
shades of meaning. He even repeated the verb tt poaayopiveiv three
times within a few lines. Is the second use of this verb— the one on
which the predeistic interpretation rests — more than an inadequate
approximation? Nothing else in the numerous notes concerning
Roman religion collected by this author, who was highly respectful
of divinity, is oriented toward predeism, equates a god with a fetfsh, or
suggests that a divine concept had been produced by animating an
object. Is this note an exception? Personally I think that Plutarch
would be disagreeably surprised if he could see the use which has
been made of his cursory remark.

The lance of Mars and, in a parallel case which is liable to a similar
refutation, the flints of Jupiter (Juppiter Lapis), are the major argu-
ments of the primitivists. Everything else that they have produced
has either been improperly interpreted or merely proves something
which is readily conceded to them anyhow, that religious acts closely
resembling magic existed at Rome side by side with religion.

As for the interpretatio romana of the Melanesian mana as numen,
nowadays it seems so inadequate, after so much severe criticism, that
its author himself declared it in a late article to be without impor-
tance, only the facts having to be taken into account and not the name
given to them. 12 He quickly forgot his recantation, however, and
until his death dauntlessly used numen in the meaning which he had
assigned to it. Others continue to do likewise. Although its use seems
to have become rarer, otherwise excellent books still refer to numen
in republican times as a "quality” possessed by a god or even by a
thing, sometimes making such references as if by mistake. It is worth-
while to sum up here a very clear dossier which has been available
for a long time, since the essentials are contained in T. Birt’s fine
article "Zu Vergil Aeneis, I, 8: quo numine laeso” (BPhW 38 [1918],
cols. 212-16), and since F. Pfister, in his monograph in Pauly-Wissowa's
Real-Encyclopctdie (17 [1937], cols. 1273-91) has contributed the best
history of the word, discounting his conclusion, where he quite un-
expectedly adopts the manaist thesis.

12. "Numen and Mana," HThR, 1951. pp. 109-30. There is no example that the domi-
nant idea of a religion has not found expression in the language of the adherents of that
religion.

“numen” or “deus”?

The dominant fact is that up to the time of Augustus, and including
Cicero, the word numen was never used alone but always with the
genitive of a divine name (Jouis, Cereris ; dei, deorum ) or, rarely, by
.ina logy, of the name of an entity or of a prestigious collective body
(mentis; senatus, populi Romani ). 13 In these phrases it does not desig-
nate a quality inherent in a god, but the expression of a particular will
of this god. Thus it conforms to its etymology. Numen is derived from
i he Indo-European root *neu-, but it was formed in Latin and not
inherited from the Indo-European (Greek vtdfia seems to be a parallel
and synonymous, but independent, formation). In Latin, then,
whatever may be its meanings in other languages, this root ( adnuere ,
itbnuere) means exclusively “to make an expressive movement of the
head and, by extension, a sign indicating approval or rejection.”
Varro (L.L. 7.85) explains it well as numen dictum a nutu, and the
beautiful supplication which Livy (7.30.20) ascribes to the Campanian
ambassadors to the Roman Senate is more than play on sounds:
mlnuite nutum numenque uestrum Campanis et iubete sperare incolumem
Capuamfuturam. Wagenvoort’s attempt to give to numen the general
meaning of “movement,” based on the meaning, in any case in-
accurate, which he attributes to the Vedic root nav-, was thus doomed
in advance. It is not saved by Lucretius (3.144: cetera pars animae . . .
ml numen mentis momenque mouetur). Here numen and momen are not
redundant; one designates the decision manifested by the sovereign
mens and the other the movement in which it is engaged (one might
ir.inslate, “the rest of the soul obeys the decisions of the intelligence
•ind follows its movements”). It is only with the Augustan writers,
ilirough the intermediary of a meaning “divine power,” which is,
mathematically speaking, as it were the integral of the god’s particu-
lar wishes, that numen becomes first a poetic synonym for “god,”
second the notation for each of the various prouinciae which form the
« omplex domain of a god, and third the expression of that which is
most mysterious in the unseen world. For an understanding of the
religion of the royal and republican periods, these developments
are without importance. Still it must be noted that Virgil often
icspects the original meaning; for example, in the verses at the

1 1. The manaists sometimes offer as evidence a fragment of Lucilius (in Nonius p. 35 L,)
tun mu men there is an emendation of nomen in the manuscripts, which is, however, accept-
able; last of all see Latte, p. 57, n. 2, and cf. H. Wagenvoort himself, Roman Dynamism ,
p 74, n. 1, in chap. 3 /'Numen."

PRELIMINARY REMARKS

beginning of the Aeneid (1.8-11), which are the point of departure for
Birt’s study :

Musa mihi causas memora, quo numine laeso
quidue dolens regina deutn tot uoluere casus
insignem pietate uirum tot adire labores
impulerit. Tantaene animis caelestibus irael

Here numine laeso can mean only “which decision manifested by
Juno having been violated” (cf. numen uiolare in Lucr. 2.614; tenter ata
uoluntas in Ov. Met. 9.267). Likewise, in the seventh book (383-84),
when the Latins decide on war against the Trojans, the poet specifies
that they do so contra omina, contra fata deum, peruerso numine, that is,
against the signs which were sent to them, against the declarations of
the gods, and acting in direct opposition to the decision expressed by
those gods (cf. Cic. Mur. 36, peruersa sententia).

To sum up, numen is unfit to fill the role to which H. J. Rose has
assigned it. Ancient usage and etymology on the contrary attest the
primacy of the concept of the personal god. Throughout the centuries
numen was only numen dei, the will expressed by such and such a god.

We must insist here on a simple fact, one which the primitivist
schools, and the predeistic one in particular, have almost succeeded
in having us forget : there is a Latin word deus, and it is no small thing.
Once more we must acknowledge the existence of Indo-European.
The word is found in the majority of Indo-European languages.
Among Indo-Iranian tongues, Vedic has devd “god,” and if in Avestan,
as the result of a profound religious reform, daeva has become the
name of the demons hostile to the gods (but still personal demons),
it has recently been proved that among the Scythians *daiva was still
the name of the gods. 14 The Gauls said devo-, the Irish dia, the Britons
de. In Old Norse, the plural tivar was one of the generic names of the
gods, along with a singular (especially in composition) -tyr. Old
Prussian had deiwas, and the Lithuanian says diewas. In Italy we know
Oscan deivai “diuae,” 15 and, very recently, Venetic %eivos 16 (in an

14. “A propos de quelques representations folkloriques des Ossetes, 2. *daiva en oss£te/'
Paideuma 7 (Festgabc H. Lomtnel) (i960) : 47-48.

15. In Umbrian deueia '* divinam” (with the force of "deorum")-

16. Hittite has (gen, SyutiaZ) "god/* which H. Pedersen has explained by *dyeu -
(see below, p. 177* n. 1): J. Friedrich, Hetkitisches Wdrterbuch (1952), pp, 194-95; cf- shvatt-
*‘day,~ Luwian Tiwat , Palaic Tiya% " Sonnefagott)” ( *dyeu 4* att-). The corresponding word
in Lydian has recently been identified, R, Gusmani, Lydisches W&rurbuch (1964), p. 92:
civ-, whence ciwali "godlike,"

"numen” or "deus”?

inscription at the museum of Vicenza, studied by Michel Lejeune).
The important fact is that wherever one can specify the meaning
*deiuo designates an individual being, personal and fully constituted,
which is what it designates in Latin (deus ; plural dim, normally reduced
to dil, di, from which diuus has been reconstructed; the accusative
plural deiuos occurs in the so-called "Duenos inscription”). The
preservation of this term is enough to destroy the predeistic con-
struction, since it proves that not only the earliest Romans but also
their Indo-European ancestors were already in possession of the type
of divinity which some persons are attempting to derive, under our
very eyes, from a concept equivalent to mana , that is, from a distorted
interpretation of this numen . Are we to admit that the ancient Romans,
having originally possessed this word with the meaning of £f personal
god,” kept it apart without using it, during a phase of reaction toward
die mana, only to restore it later, with its ancient meaning, when
1 heir new thoughts about numen-mana had given them back the
concept of personal gods? It is enough merely to state this evolution-
ary scheme to realize its absurdity. Better to admit the fact without
(orturing it: the Indo-Europeans who became the Romans preserved
without a break, without a slump, the conception which had been
already formed before their migrations and which is indicated every-
where, and has been since the dawn of history, by the phonetic
developments of *deiuos.

CHARACTERISTICS OF
THE ROMAN GODS

The other arguments of the primitivists are not pertinent to the pre-
history of any particular god, nor to the use of a technical word.
They are concerned with the general characteristics of the Roman
gods, including the most eminent ones, but particularly with the
horde of lesser divinities. Although less precise, these arguments
are more attractive, because the facts which they present are not
arguable. They offer a useful characterization of Roman religion as
compared with, for example, the Greek religions. They must,
however, be interpreted correctly.

Roughly, what does one find in the Roman world of the gods if
one removes everything that it owes to the Greeks? First, a certain
number of gods with relatively fixed outlines, separately honored
but without kinship and unmarried, without adventures or scandals,
without connections of friendship or hostility, in short, without
mythology; some, a very few, palpably the most important, are
frequently present in religious life; the rest are distributed through-
out the months of the calendar and the precincts of the city, made
real once a year by a sacrifice, but often without anyone’s knowing
(as with the gods of July) what services they can render. Around them,
in the space and time of Rome, a limitless number of points or mo-
ments proved in the past and go on proving their hidden power.
The Romans sense the presence on all sides of secret beings, jealous
of their secret, constantly on the watch, favoring or upsetting the
Romans’ undertakings, irritable or obliging, whom they themselves
do not know how to name. Finally one glimpses a number of little
groups, little teams, strictly interdependent, each member of which
is scarcely more than the name he bears — an instrumental name

3 *

CHARACTERISTICS OF THE ROMAN GODS

which imprisons him in the minor definition of a function, in an act
or a fraction of an act. In fact, these creatures scarcely appear in
literature, but the antiquaries, notably Varro, compiled lists of them
which were eagerly seized by the Christian polemists. These three
categories of beings are not even as distinct as the foregoing remarks
suggest. Some of the gods of the first group do not have any per-
sonality other than their name, which is often a collective name,
or any mode of existence other than the brief worship which is ren-
dered to them. Thus by interpenetration they give to the Roman
pantheon the appearance of a world of almost motionless shades, of a
crepuscular mass from which a few still pallid divinities have suc-
ceeded in detaching themselves, while all the others, aborted or
stunted in growth, are restricted forever to wretched and scanty
manifestations.

We cannot escape this singular impression. Accustomed as we are
to the opulent mythologies of Greece and India, of many so-called
barbaric peoples, we find it hard to imagine that Roman theology
separates at this point from fable, and that the pious souls of Rome are
satisfied with these barren nomenclatures that say nothing to the
senses and little to the spirit. There must be a basic failing here.
( )nc is tempted to admit that Roman society, which very early showed
such genius in law and politics, was stricken with an almost total
inability, in the area of religion, to create, to conceive, to explore, to
organize.

Such a deduction would be paradoxical as well as fatal for further
siudy. Before resigning oneself to it, one should proceed beyond the
impression and examine one by one the factors which contribute
10 it. Why in fact did these otherwise normal and even highly gifted
people satisfy their religious needs in these austere forms rather than
m (he rich and brilliant forms of which they were certainly as capable
.is others — as their later history proves? We shall take up, in reverse
ol der, the three elements of the divine beings, starting with the
minor Sondergotter who afforded Saint Augustine so much fun. 1

( Certain groups of these beings have remained famous, for example
those who govern the birth, the nourishment, and the course of

i . H. Lindemann, Die Sondergdtter in der Apologetik der Ciuitas Dei Augustins , Diss. Munich

PRELIMINARY REMARKS

study of the child ( Ciu . D. 4.11 and 7.3.1). After Vitumnus and Sentinus
have given him life and feeling, Opis takes him up from the bosom
of the earth, Vaticanus opens his mouth for the first wails, Levana
lifts him off the ground, Cunina cares for him in the cradle, Potina
and Educa give him drink and food respectively, and Paventinus attends
to his fears; when he goes to and returns from school, Abeona and
Adeona take charge of him under the supervision of Juno Iterduca and
Domiduca.

Consider further those beings which from hour to hour, even
from minute to minute, work in shifts and with growing indis-
cretion to facilitate the tragicomedy of the wedding night. Domidu-
cus has already led the young bride to her husband's home, Domitius
has installed her there, and Manturna has held her back until the
most delicate moment, which is described as follows (ibid. 6.9,3):

... If the man had to be helped at all costs while working at the task before
him, wouldn’t some one god or goddess be enough? Would Venus alone be
unequal to the occasion? She is said to derive her name Venus from the fact
that without violence a woman does not cease to be a virgin. If there is any
modesty among men, though there be none among the gods, when a bridal
pair believe that so many gods of both sexes are present and intent on the
operation, are they not so affected with shame that he will lose his ardour and
she increase her resistance? And surely if the goddess Virginensis is there to
undo the virgin’s girdle, the god Subigus to subject her to her husband, the
goddess Prema to keep her down when subjected so that she will not stir,
then what job does the goddess Pertunda have here? Let her blush and go
outside, let the husband also have something to do! It is surely disgraceful
for any but a husband to do the act that is her name. But perhaps she is
tolerated because she is called goddess, not god; for if she were supposed to
be masculine and so called Pertundus, the husband to defend his wife’s
chastity would require more help against him than the new mother against
Silvanus.

But Saint Augustine derived his information from serious authors
who did not dream of joking. The most remarkable list, with a
variant of which he was also aware (4.8), comes from the Libri iuris
pontificii of Fabius Pictor, through Varro, who is himself cited by the
commentator on Virgil (Serv. Georg . 1.21). This is the list of specialized
entities whose names the flamen of Ceres recites when he sacrifices
to that goddess and to Tellus (quos inuocat flamen sacrum ceriale

CHARACTERISTICS OF THE ROMAN GODS

faciens Telluri et Cereri ). 2 Each of these instrumental names ending in
-tor corresponds to a specific agricultural activity: Vervactor (for the
turning over of fallow land), Reparator or Redarator (for the prepara-
tion of the fallow land), Imporcitor (for plowing with wide furrows),
Insitor (for sowing), Obarator (for surface plowing), Occator (for harrow-
ing), Sarritor (for weeding), Subruncinator (for thinning out), Messor
(for harvesting), Connector (for carting), Conditor (for storing), and
Promitor (for distribution). In a neighboring sphere, when the Fratres
A r vales perform certain expiatory rites for the trees which they
remove from the Incus of Dea Dia, they break up the movement and
address themselves separately, as we know from their own Acta,
to Addenda , Commolenda , and Deferunda (in a.d. 183), to Addenda and
Coinquenda (in 224); these four entities, restored from alphabetical
order to the natural order of their interventions, had as their provinces
the removal of the tree, parceling it out, cutting it up, and burning it. 3

In these classic examples the primitivists see the earliest form of the
divine representations of the Romans. According to them, the best-
characterized gods emerged from groups of this type into a some-
what wider or more important sphere of activity, or by a process of
imitating Greek models. This theory is completely mistaken. On
one hand, the specialist entities, who always appear as a team, and
on the other hand, the autonomous divinities, form two irreducible
categories, answering to different needs. It is not possible to produce a
single example, in the historical era during which these lists are
numerous, of the kind of promotion by which a divinity on the level
of Vervactor or Pertunda became a divinity such as Ceres or Juno.
Nothing allows us to think that it was otherwise in prehistory. The
catalogs of indigitamenta — the name of these litanies — are not breeding
places for the gods, and the entities gathered together in these catalogs
remain there.

2. On this series, see J. Bayet, "Les Feriae Sementiuae et les indignations dans le culte
* le ( ^eres et de Tellus,” R HR 137 (1950) : 172.-206, esp. p. 180.

l. See references and commentary in Latte, p. 54, who rightly points out that neither
ihc names nor the rites can be ancient. But if there was an archaic model for the names,
we are not obliged, as Wagenvoort and Latte believe, to translate "[the tree] which must
be removed [cut up, etc.]/* These names of pseudo-divinities were fabricated with great
lieedom, and Deferunda may simply be, with an animate feminine ending, the neuter
impersonal deferundum "it is necessary to remove.” Moreover, if one insists on attributing
ihrso words to the archaic language, one should not forget that even in Latin the forma-
i ions in -n-d- had a very wide meaning, not necessarily passive, E. Benveniste, Origines de la
formation des noms en indo-europien 1 (1935): 135-43.

PRELIMINARY REMARKS

A few remarks must be made concerning these groups. First, the
conceptual zone in which one finds them is limited. The known
indigitamenta are related only to rural operations and to private life.
The activity of war, for example, in which one might easily imagine a
great proliferation of functions relating to the handling of weapons
or to field maneuvers, does not contain any such. Whereas the flamen
of Ceres calls on Vervactor and his companions, his colleague, ^he
flamen of Jupiter, whose religious activity is constant and about whom
we know a great deal, does not have to manage any such teams.

Second, these entities certainly do not have any great importance,
even in the limited areas where they are active. Vervactor and his
companions seem not to have had any activity, and hence no existence,
outside of the flamen’s prayer. Cato, to whom we owe our knowledge
of so many rural rites, does not mention them, nor does any author,
not even the poet of the Georgies, or the author of the Fasti , nurtured
on Varronian scholarship, or the Cicero of the theological treatises.
Varro himself, who collected them in his religious books, does not
have occasion to speak of them in his book on agriculture. Long
before they amused Saint Augustine, the entities of the wedding
night might have provided the comic and satiric poets with good
effects; Plautus and Juvenal ignore them, however, as they ignore
every other enumeration of this sort. Everything takes place as if these
roll calls had remained the province of a few specialists in sacrifice,
having no more influence on the religion of other Romans than the
angelic hierarchy has on the devotion of the average Catholic, which
the different Prefaces of the Mass give in several variant forms:
Angeli atque Archangeli , Cherubim quoque ac Seraphim ; or, cum Angelis
et Archangelis, cum Thronis et Dominationibus , cumque omni militia
coelestis exercitus ; or, per quern maiestatem tuam laudant Angeli , adorant
Dominationes , tremunt Potestates , Caeli caelorumque Virtutes ac beata
Seraphim soda exsultatione concelebrant. The principle of classification
is entirely different in the indigitamenta and in the Prefaces, but in
both cases the enumeration remains the concern of the Iiturgists and
the theologians; it does not pertain to the living religion.

In the third place, it is possible that these lists were modeled on
certain earlier ones, which were perhaps different from those which
we now have. It is notable that what we know of the old song of the
Fratres Arvales, the priests who are concerned with the fields, does

CHARACTERISTICS OF THE ROMAN GODS

not mention the team of Vervactor but only the great gods who de-
fend or house the grain, or “work” it from inside: Mars, the Lares,
and the Semones.

Finally, the best-known lists of indigitamenta always appear in a
subordinate position. Not only are the entities which figure there
never the objects of a cult, the special titulars of a priest, but it is the
II amen of Ceres who invokes Vervactor and his companions, when he
sacrifices to Ceres and Tellus. What can we say but that they belong to
( ^cres, that they perform minor details and act subordinate roles in
i lie whole sphere of her work, which is to further the prosperity of the
fields? Similarly, among the entities concerned with the child, ac-
cording to a formal statement by Saint Augustine ( Cm . D. 7.3.1) Abe-
011a and Adeona, deae ignobilissimae , are associated in their duties with
Juno, the “select queen of the gods, the sister and wife of Jupiter.”
What we know of Juno, the protectress of marriages and of childbirth,
suggests that it is from her that the series Virginiensis, Subigus, etc.,
and the series Opis, Vaticanus, etc., derive their functions.

These few remarks suggest a reasonable coexistence of the gods,
properly so called, and the lists of indigitamenta . The minor entities
grouped in one of these lists are as it were the familia of a “great” god.

I can only repeat what I proposed on this subject sixteen years ago. 4

At Rome as elsewhere, in order to understand the society of the gods, we
must not lose sight of the society of men. What do the private life and the
public life of the Romans teach us here?

In private life, let us think of those great gentes who have slaves by the
1 housands, Jamifra rustica, familia urbana , the majority of whom have special-
ized duties, as the pistor, the obsonator , etc. Let us think of the comic enumera-
lions like that in the Miles Gloriosus (693-98), where a husband complains of
1 he demands made by his wife during the Roman festival of Quinquatrus (Da
ifitod dem quinquatrubusl ): she has to give money to all the different kinds of
witches, praecantatrici, coniectricU hariolae , haruspicae y piatrici , as well as to her
specialized servants, the ceraria , the obstetrix , the nutrix of the uemae — and
1 he Roman spectators were amused by this picturesque ~ roll call,” as they were
by the catalog in the Aulularia, where it is no longer the famuli who are listed
m eleven lines, but the artisans, gathered in the service and at the whim of the
g'eat ladies (508-22): statfullo , phrygio, aurifex , lanarius , caupones y patagiarii ,
indusiariiyflammariiy uiolarii , cararii y propolae , linteones , calceolarii y etc.

1 DIE, pp. 122-25 . 1 am grateful to Les Presses Universitaires de France for authorization
10 reproduce this extract and the one on pp. 45-46.

PRELIMINARY REMARKS

As for the public life of Rome, let us think of all the apparitores , the lictor ,
the praeco, the scriba , the pullarius , who accompany the high magistrate, each
with his particular competence; and of the priests calator , the candidate's
nomenclator , etc.

In such a society, which loved lists, specifications, method, and well-divided
work, it was natural that the true gods, great and small, should also have
within their prouincia auxiliaries named for the single act which each had to
perform, under the responsibility and for the advantage of the god. Onqe the
mold was established and the habit formed, this kind of personage multiplied
in every area, with popular names, sometimes with names just as "poorly
made” or with an etymology as approximative as certain products put out
nowadays by our drug manufacturers.

We are not speaking here of a "primitive,” or even of an inferior form of
religion. It is on a second plane, indefinitely peopled, joined to the first. So
much the more, I repeat, these entities are not candidates for the rank of
"great gods.” The paterfamilias cannot be the "transformation” of the pistor
or of any other -tor specialist living in his house or on his farm. The consul
cannot emerge from the level of his followers. On the contrary, it is the miller-
slave, the provisions-slave, the wheelwright-slave, etc., who require the
existence of the paterfamilias and depend on him ; and it is the lictor , the quaes-
tor, etc., who imply the province, the will, and the person of the consul.
Similarly Ceres is not a chance variety of Vervactor, Repara tor, etc., who has
emerged from the mass; quite on the contrary, it is these minor personages
who imply the preexistence, on a level above theirs, and with the widest
competence and resources, of Ceres.

And are we to be astonished that the Romans knew nothing particular or
personal about Vervactor, for example, outside of his one act? We might just
as well be astonished that we know nothing, not even his name, of the lictor
to whom Brutus turned over his sons, or of the lictores to whom Manlius or
the cruel conquerors of Capua said, anonymously, "7, lictor , deliga adpaluml ”
In the eyes of the citizen this functionary presents, in political life, exactly the
same very limited angle of interest as Vervactor and his companions in the cult
of Ceres. Let us not conclude that the Romans were incapable of going beyond
a schematic type of divine person. Let us rather conclude, as we replace things
in their natural historical order and rely on the ordinary parallelism between
the invisible and the visible, that they were sparing of exact specifications, and
that their imagination, while it multiplied the specialized auxiliaries of the
gods in imitation of their social life, did, not waste its time in conceiving and
saying more about them than was necessary.*

5. As things are, we must not expect clarification from the etymology of the words
nouensiles and indigetes. E. Vetter, "Di Novensides, di Indigetes," IF 62 (1956): 1-32, is not
convincing.

CHARACTERISTICS OF THE ROMAN GODS

The primitivists advance other kinds of facts which seem to them to
show how hard it was for the Romans to imagine a god, to do more
than sense his presence, or to liberate him from the physical phenom-
ena in which he manifested himself.

First there is the frequent uncertainty concerning the sex of the
divinities. This the Romans accepted, and it did not seem to disturb
them. Liturgical formulas, preserved by the historians and imitated
by the poets, used the phrase siue deus sine dea. According to a note
by Servius ( Aen . 2.451), a buckler was preserved at the Capitol,
bearing the inscription Genio Romae , siue mas siuefemina . Furthermore,
the Romans did not even know whether the really ancient divinity
named Pales, who was important both in the rural economy and in
national history, was a god or a goddess, and opinions on this subject
were divided.

An even greater uncertainty is revealed with regard to a number of
local or temporal entities. 6 When a Roman said genius loci , he said
everything he could about the kind of supernatural being which
revealed itself indirectly, by a single phenomenon, and in a fixed
place; and he was resigned to this paucity of information. The speci-
fications of the concept of Fortuna seem to be particularly convincing.
Fortuna huius did (Cic. Leg . 2.28), to whom a temple was vowed
during the battle of Vercelli in which the Cimbri were defeated, is
analogous in her inconsistency to Fortuna uirilis , Fortuna muliebris y
and many others, whose sanctuaries are found throughout Rome.
Is not the limit of Roman creativity in the matter of gods, the in-
capacity of the Romans to flesh out and materialize a being that they
do not see, conveyed in this expression which claims to be adequate?

There is the same lesson, we are told, in the designations of gods
such as Aius Locutius: once a voice spoke to the Romans, with predic-
lions of dire events in the near future. In this voice, to be sure, they
recognized an ally from the beyond, but all they could manage to
< < inceive of it was expressed in a double name derived from two verbs
1 1 leaning “to say”; they knew that it had spoken, and that was all.
What would not the fertile imagination of the Greeks or the Indians
have invented under similar circumstances!

These facts are true on the whole, but perhaps they are presented

<>. K. Latte, " Uber eine Eigentumlichkeit der italischen Gottesvorstellung,” ARW 24
( 244-58.

PRELIMINARY REMARKS

somewhat tendentiously, and above all they find a much better
explanation outside of primitivism. The picture presented here is
not one of complete impotence. The Roman was certainly able to
determine the sex of a divinity, and, like the Greek or the Indian, he
could make up stories. If he does not do so, it is by virtue of a character,
a quality of spirit which is not peculiar to him, to be sure, but which he
has developed to an astonishing degree — prudence. His magistrates,
who are all professional jurists, know the importance of formula^ and
the necessity of not exceeding by one syllable the inventory of known
facts in the relations between gods and men as well as those between
man and man. Others have used words to produce lyric poetry, have
dreamed, have amplified; the Roman, on the other hand, for many
years used his language to compose exact and useful statements, and
the elegance of this literary genre is not that of poetry.

Throughout this book we shall find this juridical spirit pervading
the worship and the whole fabric of religious life with its reservations
and its precisions; it occurs in a great variety of forms, many of which
are specific illustrations of prudence. In legal chicanery, in procedure,
in the efforts of the lawyer as well as in the wording of claims, one
must be cautus. The very word for a lawsuit and its contents, causa ,
is doubtless derived from this root, and the magistrate who had to
condemn men to death said simply, “male cauerunt.” The word
which finally designated the whole of man’s relationship with the
invisible, religiones, religio , whatever its etymology, originally expressed
caution: not a flight of the spirit, or any form of activity, but a halt,
the uncertain hesitation in the face of a manifestation which one must
first understand fully before one adjusts to it. The Romans thus very
quickly measured the force and efficacy of words in the field of
religion. One of the first "myths” that one reads in the vulgate of the
beginnings tells of a bargain between Jupiter and Numa, which is at
the same time a test whereby the god makes sure that the king
understands the importance of vocabulary and syntax. The god
expresses himself poorly, giving his opponent an opening. "Cut off
a head!” he says. Numa replies, interrupting him with the words
. . an onion.” "A man’s,” the god rejoins, without specifying pre-
cisely that he wants the human head which he first mentioned.
Numa seizes his second chance, and again interrupts with “his
hair.” The god repeats his mistake once more, and demands "a

CHARACTERISTICS OF THE HOMAN GODS

life”; "of a fish,” interposes Numa. Jupiter is convinced; he rewards
and as it were confers a diploma on this brilliant student. "See to it,”
he says, "that by these things thou dost expiate my bolts, O man
whom none may keep from converse with me [o uir conloquio non
abigende meo]” 7 (Ov. F. 3*33^-44).

Such is the constant care of the Roman in his religious as well as his
social life: to speak without imprudence, to say nothing and above
all to use no formula by which the god or his human spokesman
might benefit at his expense, nothing which might irritate the god,
or which might be misconstrued or misunderstood. No less legendary
ilian the colloquy of Jupiter and Numa but no less revealing is the
famous account of the events which guaranteed to Rome the privilege
( >f the Capitoline omen. While excavating to establish the foundations
of the temple of Jupiter Optimus Maximus, the diggers brought to
light a mans head. The Romans sent emissaries to Etruria to consult
a famous soothsayer on the meaning they should give to this dis-
covery. They did not suspect the risk they were running. By a clever
play on the two adverbs of place illic and hie , "over there ” and "here,”
i he Etruscan would surely have dispossessed and robbed them of the
omen and of its meaning, if, by a providential betrayal, his own son
had not warned his guests (Dion. 4.60):

I [ear me, Romans. My father will interpret this prodigy to you and will
tell you no untruth, since it is not right for a soothsayer to speak falsely; but,
tn order that you may be guilty of no error or falsehood in what you say or
in the answers you give to his questions (for it is of importance to you to know
these things beforehand), be instructed by me. After you have related the
prodigy to him he will tell you that he does not fully understand what you
\ay and will circumscribe With his staff some piece of ground or other; then he
will say to you : "This is the Tarpeian Hill, and this is the part of it that faces the
east, this the part that faces the west, this point is north and the opposite is
south.” These parts he will point out to you with his staff and then ask you in
which of these parts the head was found. What answer, therefore, do I advise
v ou to make? Do not admit that the prodigy was found in any of these places
hr shall inquire about when he points them out with his staff, but say that it
.1 1 ipeared among you at Rome on the Tarpeian Hill. If you stick to these
answers and do not allow yourselves to be misled by him, he, well knowing
ih.it fate cannot be changed, will interpret to you without concealment what
1 lie prodigy means.

/. Var. deumfor tm>.

PRELIMINARY REMARKS

And what a meaning! The soothsayer is finally obliged to reveal it:
the place where the head was found will be the head of Italy.

The formulas of the fetial priests at the conclusion of a treaty, as
Livy has transmitted them (1.24.7-8), bear witness to the same con-
cern for precision. Before he sacrifices the pig to Jupiter, the guarantor
of treaties, the priest declares that the Roman people will not be the
first to violate the clauses uti ea hie hodie rectissime intellecta sunt , and
that if they do, Jupiter will be able, on that day, ilia die , to skrike
them sicut ego huncporcum hie hodie feriam.

It is in this tradition of verbal prudence that we must interpret
limitative specifications like Fortuna huius diei , huius loci , etc. They are
explained by the Romans' concern to circumscribe in space or in
time what is expected, hoped for, or feared. Not that the Romans
had not for a long time had the concept of fortuna. But as a general
term it was of no use to them, since they were not greatly concerned
with a long-range, universally valid destiny. Therefore they made the
idea specific according to its places and times of application. Had such
a place or such a day, in their experience, shown itself to be particu-
larly propitious or harmful to Roman undertakings? Then they might
reasonably conclude that some variety of fortuna would in the future
require the particular grateful or precautionary attention of men. For
example, the Allia had been the site of a cruel defeat and one heavy
with consequences, at the time of the Gallic invasion. Thus, at a
later date the Romans hesitated to engage in another battle there —
which is what Rome’s enemies, the Praenestines, were hoping for.
It was not that they feared fortune in general, or the fortuna of battle,
or that of the consul, or that of the day, but the specific, circumscribed
fortuna of the place (Liv. 6.28.7 and 29. i). 8

This attitude is found in all religions in which formulas are used.
The poets of the QgVeda, whose learnedly daring lyricism we may
contrast with the matter-of-fact spirit of the Romans, do not escape
from it when the occasion demands it. Do they seek the protection
of Indra (8.61.16-17)? " Protect us behind, below, above, before, on all
sides, O Indra! Ward off from us the danger which comes from the
gods, ward off the blows of the non-gods. Protect us, O Indra, every
today and every tomorrow; protect us always, day and night!”
The caution here expressed consists in a complete summation, not in

8. See the commentary in DL , p. 79, n. 3*

CHARACTERISTICS OF THE ROMAN GODS

,1 thorough limitation, but the intention and the need are the same :
(o express oneself in a way that leaves the being whom one is address-
ing no loophole. The AtharvaVeda provides a number of similar
examples, and one might cite, from the myths of India, Ireland, and
elsewhere, more than one misfortune which befell gods or demons
because they mispronounced an important word or put the accent
in the wrong place, or because they delimited inadequately the “yes”
with which they answered a request. But these displays of prudence
are particularly evident in Roman worship, perhaps partly because we
know only their formulary framework*

[t is certainly still prudence which explains the formulas of the
lype sine dens siue dea? or the even less ambitious designations like
Aius Locutius.

With regard to the former, we must first clear up a misapprehen-
sion. It is not correct to say that the Romans ever agreed not to give
a sex to a specific divinity whom they knew in other respects and to
whom they had given a name. As we shall see, the example of Pales,
which is always cited here, does not hold out against the rigorous test
of the facts. Pales, or rather the two Pales, of Rome are both goddesses,
and the masculine Pales is found only in Etruria; this means merely
i hat the Etruscan divinity with the same function as Pales was a god,
not a goddess. At Rome, however, neither the shepherds nor the
scholars were misled by this . 10 When one reads on the buckler at the
( apitol Genio Romae , siue mas siuefemina, the word siue , here and in all

«>. The opening of the formula of euocatio is often dted at this point (Macr. 3.9.7). It is a
I . il.se interpretation: the context proves that si deus , si dea est cui populus ciuitasque Cartha-
vwicnsis est in tutela means “ all the gods and goddesses who protect the people and the

• iiy of Carthage/’

io. Below, pp. 380-84. The other “hesitations about sex” that are cited are not that.

• Horns does not exist at Rome beside Flora; it is only in Oscan territory that he is found,
.iiul even that is not certain (Vetter, p, 183). *Pomo or *Pomonus is Umbrian (Tig. Ill 26.

« i( .; cf. A. Ernout, Aspects du vocabulaire latin [1954], p. 29, n. 1) and does not coexist in
I .Kin with Pomona (the Pomones, a lexical term [Gloss. Ansileubi : Gloss. Lat. 1 (1926): 450]

• Mined as “pomorum custodes” could only be “small change,” in the plural, of Pomona ,
l< >* 1 He various kinds of poma; word modeled on Semones ?). Tellumo is not the masculine
doublet of Tellus, but a name of indigitation in the series Tellumo Alter Rusor (Aug. Ciu. D.

• ; \.i).Jana is more than doubtful. If there ever was a goddess Coca, which has not been
rsublished, Cacus is solely a character in legend. In fact, there is a coupling (and no hesi-
1 •*! ton !) only of the divinities who govern sexuality ( Liber-Libera ) or are heavily involved
m .sexuality (Faunus-Fauna). Obviously such pairs as Quirinus and Hora Quirini or of the type
« 'I ft bus and fer/us Martins (Iguvium) do not at all prove that the Romans or the Umbrians
li.id trouble in distinguishing the second divinity from the first.

PRELIMINARY REMARKS

analogous cases, is intended to prevent confusion and ambiguity.
The Romans know that there is, somewhere, a divine being, but they
have no further information; and while trying to make clear their
meaning so that it cannot possibly be misconstrued, they are unwill-
ing to exceed the limits of their knowledge. The idea of a feminine
Genius is strange, and for this reason, the information given by Servius
is suspect. But the thought process which justifies the phrase is en-
tirely real and Roman. If, for instance, they had said "he,” if they liad
been satisfied with a masculine Genius , and it turned out to be "she,”
then "she” would have had entire freedom to pretend that she ig-
nored the offering of the buckler, or even to take offense. In sum, this
is only a particular instance of a very general attitude. In the same
passage Servius cites a prayer formula: et pontijices ita precabantur
" Juppiter Optime Maxime, siue quo alio nomine te appellari uolueris . .
The pontiffs certainly conceive of Jupiter precisely, and as fully as a
Greek conceives of Zeus, and they are aware that his principal
epithets are Optimus and Maximus. But they reserve the possibility of
incomplete information, of some caprice on the part of the god, of
any uncertainty, and they provide for these contingencies in the words
which they add to the epithets. The fifth book of the Aeneid (94-96)
offers a good example of the same circumspection. At the end of the
year Aeneas has placed an offering on his father's tomb. A serpent
appears, tastes the food, and then goes back into the tomb. Aeneas
does not know what this animal is — the genius of the place? a
messenger from his father? incertus Geniumne loci famulumne parentis
esse putet. And so he repeats his offering. Shall we conclude that
Aeneas, which is to say every Roman, is incapable of distinguishing the
different kinds of spirits? On the contrary, it is because he is familiar
with these varieties, and because he does not know which one the
serpent represents, since it is multivalent, that he repeats the rite. The
uncertainty exists and is felt by the celebrant, but it results from an in-
sufficiency of information; he is aware of this and prudently bases his
conduct on it. 11 The expression siue deus siue dea does not prove that
the Roman had difficulty in imagining sexually differentiated gods;
it proves merely that in a specific case he does not feel himself to be
sufficiently informed and that he prefers to envisage both possibilities.

11. J. Bayet, "Les cendres d'Anchise: dieu, hdros ou serpent?” Gedenkschrift G. Rohde
(1961), pp. 3^-56.

CHARACTERISTICS OF THE ROMAN GODS 45

The modern businessman who prints up ten thousand copies of an
advertising circular beginning “Dear Sir or Madam” operates in the
same way.

The case of Aius Locutius can be explained in the same way; it does
not reveal a weakness of conception, nor is it an extension of the
primitive:

The Romans knew that a god had intervened in a famous occurrence, since
he had spoken, and his voice had been heard ; they even knew that it was a
male god, since it is possible to recognize a person’s sex by his voice. But they
knew nothing more about their benefactor. To be sure, they might have
expressed their certainties and circumvented their lack of information by
using an analytic formula, by saying, for example, “the god who spoke.” This
was too long. They said, “Aius Locutius ” But the intention and the result are
the same. Can it really be disputed that the Romans, if such had been their
wish, could have imagined the god who had spoken to them, in human form,
complete with features and expression, or that they would have made con-
jectures regarding his identity with one or another of the known and honored
gods? They simply did not want to. Their experience in lawsuits, in this legal
art which is essentially precise and cautious, had taught them that it is better
not to add imaginary and unverified elements to the proven data of a file,
even if the latter are very limited. Artificial speculations open the way to
risks, and especially to the risk of misdirection. In the most favorable cases,
where the marvelous intervention was somehow identified, they did not
hesitate over the god’s nomen, and merely gave him a new cognomen derived
from the event. For instance, in all the cases of uota, the Romans were sure
that the power who had halted the panic on a certain battlefield, or had
achieved the victory on another, was Jupiter, since it was to Jupiter that the
general had vowed a temple in order to obtain this success. Thus the worship
and gratitude of the Romans were naturally addressed to Jupiter, Stator or
Victor. Doubtless in the same way a specific detail, later forgotten and re-
rcplaced by various mutually incompatible legends, guaranteed that the
goddess to whom they owed a certain piece of precious advice was Juno.
Therefore they worshiped not some vague “Nuntia Moneta ” but, with com-
plete assurance, Juno Moneta — whom a picturesque and unpredictable future
was to make the eponym of our “money.” The Romans could just as easily
have formed the flattering hypothesis that the god who had spoken in a
Roman street on the eve of the Gallic invasion was Jupiter, and called him
Jupiter Locutius, as they spoke of Jupiter Elicius. But this would have been
dangerous ; it might have irritated Jupiter, if by chance he had had nothing to
do with the incident, as well as the god who actually had spoken. Therefore

PRELIMINARY REMARKS

4 6

they merely used a double expression, “Aius Locutius,” for which there were
some precedents. This was not the impotence of savages, but once more, as
always, the prudence and verbal caution of a people experienced in legal
procedure . 12

12. DIE, pp. 138-39* The same arguments would be made, for example, for the “god”
Rediculus, who had a fanum extra portam Capenam. He owed his name to the fact that
Hannibal, after coming very close to Rome, ex eo loco redierit (Fest., Paul. pp. 384-85 L 2 ).
At the time of the Second Punic War the Romans were certainly capable of conceiving of
gods with a complete personality; but, not knowing which god had inspired the/visions
that persuaded Hannibal to retreat ( quibusdam perterritus uisis), they chose this prudent
way out.

THE LOST MYTHOLOGY:
THE EXAMPLE OF THE

MATRALIA

There remains the great peculiarity of the Roman gods, this depriva-
tion of all mythology and all kinship, this “ inhumanity, ” and, in
many cases, outlines and activities so vague that in the end we know
less about them than about Vervactor, Pertunda, or Aius Locutius.

This last point is only too well established: a goddess like Furrina,
important enough to be the titular of one of the twelve minor flamens,
remains almost as mysterious to us as she was to Cicero, who was
able to say something about her only by means of an etymological
pun (furid). She is not the only one. Here too a few remarks will
suffice to remove the argument which the primitivists base on such
divinities.

First, the number of these mysteries has been appreciably reduced
during the last twenty-five years. Of the list of enigmatic figures which
Wissowa drew up, the majority have been given a plausible meaning,
in which all the items of the file fall into place harmoniously and which
.ire supported by parallel cases found among other Indo-European
peoples. Among the goddesses, not to revert to Pales, whose entire
mystery consisted in the uncertainty of her sex, Carna, Diva Angerona
together with Volupia, Feronia, and even Lua Mater can now be
understood; in their special functions they are peculiar, but as divin-
ities they are normal, even banal . 1 Among the male gods, the classic
example of Quirinus is not more tenable; he too can be understood
in his own right and in his relations with the two other gods of the
major flamens, both in his original form and in his later development.
To obtain these results it has been sufficient to refrain from strong

i. See below, pp. 380-84 (Pales), 385-87 (Carna), 335-37 (Angerona), 414-21 (Feronia);
Ibr Lua, DL, chap. 4.

PRELIMINARY REMARKS

4»

prejudices, as, for example, in the case of the goddesses, from the
barren obsession with the “Mother Goddess” or the “Great Goddess,”
and, in the case of the god, the Sabine illusion and a few others, about
which we shall speak again shortly.

This first comment leads naturally to another. The imprecision, the
uncertainty which was thought to be congenital among the divinities,
does not in fact exist. To be sure, the Romans of the classical period
misunderstood Quirinus, and if they left good definitions of C&rna,
they had strong doubts about Angerona. Their hesitations over both
Quirinus and Angerona have contributed not a little to complicating
the file and making its investigation difficult for us moderns. But this
annoying situation does not mean that Quirinus and Angerona were
poorly formed, incomplete, and at all times uncertain divinities. On
the contrary, our uncertainty is the result of a weakening, an aging
produced by the forgetting of definitions and functions which had
earlier been clear, complex, and harmonious. In the era when it was
fixed in literature, a great part of the Roman pantheon was on its
way to dissolution. The Greek flood had submerged everything
and had destroyed the taste for and awareness of traditional explana-
tions. The most original forms, those which had not been able to
receive an interpretatio graeca , were destined to disappear or to survive
only in rites which became less and less intelligible. In the face of this
disaster, it is a wonder that the four or five disparate facts survived
which, together, clarify Angerona, or the two mentions which give
some meaning to Carna. In the case of Furrina this miracle simply did
not occur. In short, none of these divine personalities whose vagueness
disconcerted the antiquaries of the dying Republic and of the Empire
before they baffled modern exegetes suggests any “primitive” form
of religion.

But it is true : not even the most important and most vital gods have
any mythology. Take for example the gods of the major flamens,
Mars and Jupiter, who, without being confused with each other, work
together in the growth of Rome, and* Quirinus, who is defined as the
opposite of Mars and yet, paradoxically, is sometimes confused with
him. These gods do not take part in any adventures, either all three
together or any two of them, not even dividing the prouinciae of the
world and of the state in a way comparable to what Zeus and his
two brothers did after their defeat of the Titans* Outside of the

THE LOST MYTHOLOGY

sacra which are offered to him, the auspicia which he sends, and the
thunderbolt which he governs, all that the Romans know about the
earliest Jupiter amounts only to the promises which he made to
Romulus and Numa and the punishment which he inflicted on the
imprudent Tullus. Still, in these few cases, the great god's partner is
a man and his actions are part of “history." At the end of the Etruscan
period, when the three Capitoline gods were united in a single temple
with three cellae, it is not certain that Juno Regina was Jupiter’s “wife,"
even though an indirect Greek influence was probable about this time.
( )ps and Consus form a theological and ritual couple, not a married
pair.

Dionysius of Halicarnassus expressed the admiring astonishment of
his compatriot philosophers at this absence of fables. Convinced that
i he Romans are Greeks, through the intermediacy of the Albans, he
celebrates with complete freedom the wisdom of the earliest institu-
i ions and the religious purity of these settlers of the Occident. Romu-
lus, he thinks, copied “the best customs in use among the Greeks,"
but he knew how to limit his borrowing from them (2. 18-20):

. . . But he rejected all the traditional myths concerning the gods that con-
1.1 in blasphemies or calumnies against them, looking upon these as wicked,
useless and indecent, and unworthy, not only of the gods, but even of good
men; and he accustomed people both to think and to speak the best of the
rods and to attribute to them no conduct unworthy of their blessed nature.

Indeed, there is no tradition among the Romans either of Ouranos being
< astrated by his own sons or of Kronos destroying his own offspring to secure
himself from their attempts or of Zeus dethroning Kronos and confining
his own father in the dungeon of Tartarus, or, indeed, of wars, wounds, or
bonds of the gods, or of their servitude among men

Let no one imagine, however, that I am not sensible that some of the
« h eck myths are useful to mankind, part of them explaining, as they do, the
works of Nature by allegories, others being designed as a consolation for
Inn nan misfortunes, some freeing the mind of its agitations and terrors and
« lea ring away unsound opinions, and others invented for some other useful
purpose. But, though I am as well acquainted as anyone with these matters,
nevertheless my attitude toward the myths is one of caution, and 1 am more
inclined to accept the theology of the Romans, when I consider that the ad-
vantages from the Greek myths are slight and cannot be of profit to many,
but only to those who have examined the end for which they are designed;
and this philosophic attitude is shared by few. The great multitude.

PRELIMINARY REMARKS

unacquainted with philosophy, are prone to take these stories about the gods
in the worse sense and to fall into one of two errors: they either despise the
gods as buffeted by many misfortunes, or else refrain from none of the
most shameful and lawless deeds when they see them attributed to the gods.

We now have direct proof that this bareness, this insensibility of the
Roman pantheon, is not in itself primitive. Like all the other Indo-
European peoples, the Romans at first loaded their gods witty myths
and based their cultic scenarios on the behavior or the adventures of
the gods. 2 Then they forgot all that. It sometimes happens, however,
that we can discern the myths through the characteristic marks they
left on the rites which they originally justified and which, after their
disappearance, became insoluble puzzles, even for the Romans of the
great era. The myths which we can thus recover are in strict harmony
with stories told by the Vedic Indians and, sometimes, by the Scandi-
navians. I shall give only one example, but in some detail. 3 Others
will be indicated later in the book.

On ii June the Romans celebrated the Matralia, the feast of the
goddess Mater Matuta. 4 Despite many discussions, despite prodigies
of ingenuity expended to obfuscate what is perfectly clear, Mater
Matuta is Dawn. Matuta is the name from which the adjective matu-
tinus was formed, and in this kind of derivation the adjective never
adds anything fundamental to the substantive. To be sure, the name
Matuta belongs to a large family of words ( manus “good/* maturus
"ripe,” etc.) whose common element is the idea of "being just in
time”; but each of these words has followed its own line of develop-
ment, and Matuta came to be the deified name for the "break of the
day.” It was understood in this way by the ancients (Lucr. 5.650). 5

2. Despite H. J, Rose, “Myth and Ritual- in Classical Civilisation,” Mnem., 4th ser., 3
(1950) : 281-87, who denies the existence of myths at Rome.

3. What follows sums up chap. 1 of DL, completed and corrected — for the second rite
— on the basis of useful criticisms (A. Brelich, J. Brough) which I regret having originally
underestimated. Discussions of earlier exegeses will be found there, notably that of H. J.
Rose, based on an extravagant use of sororiare , which he connects with the root of the Ger-
man schweflnt. One is sorry to find this again in Latte (pp. 97, n. 3, and 133: Juno Sororia
as goddess of puberty!); see DL, pp. 14 (and n. i)-i6, and Aspekte, p. 21, n. 6.

4. The Matralia are nothing else but this festival of Mater Matuta, and there is no way
of distinguishing them; contrary opinion of A. Ernout, RPh. 32 (1958): 151*

5. I sum up here the demonstration in DL, pp. 17-19; cf. M. Pokrovskij, “Maturus,
Matuta, matutinus, manus (manis), manes, mane,” KZ 35 (1897): 233-37. Latte (p. 97)
pays no attention to this, and prefers to explain Matuta, in which he refuses to see Dawn,

THE LOST MYTHOLOGY

By chance we know two very precise rites of the Matralia, her feast,
which is restricted to ladies, bonae matres (Ov. F. 6475), who have been
married once, uniuirae (Tert. Monog . 17). Plutarch, confirmed by
Ovid for the second (F. 6.559, 561) and partially for the first (ibid.
ssi-58), refers to these rites several times (Cam. 5.2; Q.R. 16 and 17;
lor the second rite, Mor. 492D = De frat. am., end of 21), These
rites are as follows: (1) While the temple of Matuta is normally for-
bidden to the servile class, the ladies assembled for the feast bring
into the enclosure a slave woman, whom they then drive out with
slaps and blows; (2) the ladies bear in their arms, "treat with re-
spect,” and commend to the goddess not their own children, but those
of their sisters. These actions had to follow one another in this very
order, as it appears in the two references by Plutarch. The Romans
iliemselves did not comment on them, and they have provoked
modern exegetes to many fantastic interpretations. A glance at the
I )a wn goddess of the Vedic Indians, U$ds, allows us to understand them
<<nnpletely: what the Roman ladies do once a year, at the Matralia,
is done every morning by U$a$, or by the Dawns, Usdsah , in a collec-
livity which the poets sometimes mobilize for each particular morn-
ing, namely, a negative service of cleaning, and its positive corollary.

1. Dawn "chases the black shapelessness” (bddhate kr$ndm dbhvam :
KV 1.92.5), "drives back the hostility, the shadows” (apa dveso
bddhamand tdmamsi: 5.80.5), "drives back the shadow [dpa bddhate
idmah] as a heroic archer chases his enemies” (6.64.3); the Dawns
hold off and pursue the shadow of night [vi td badhante tdma urmya-
vtth\ while directing the head of the high sacrifice” (6.65.2); "Dawn
1 lie goddess marches, driving back [bddhamdna] by her light all the
shadows, the dangers; here the brilliant Dawns have shown them-
selves; . . . the shadow has gone to the west, the displeasing one”

l'\ 1 Ik* Oscan Maatuis Kerriiuis (dative) in the ritual of the Table of Agnone (Vetter, no.
1 1 \ line 10). These, being "Cereal/’ should rather be compared with maturus. Is it not
1 nin e sensible to listen to Lucretius :

Tempore item certo roseam Matwta per or as
aetheris auroram refert et lumina pandit . , ,

I lin e is at Rome no ancient religious reference to the Dawn other than as Mater Matuta
nid her festival. It is possible, however, that she remained alive in folklore. She had her
m venge later. R. Rebuffat, "Les divinites du jour naissant sur la cuirasse d’Auguste de
1*1 mu Porta,” MEFR 73 (1961): 161-228; “Images pompSiennes de la Nuit et de L’Aurore,”
Wl I K 76 (1964): 91-104; “La relive des bergers chez les Lestrygons,” MEFR 77 (1965):

IM .|8.

PRELIMINARY REMARKS

(i ap&dnam tdmo agdd djustam) (7.78.2 and 3). The hymns thus portray
the natural phenomenon of the coming of day as the violent driving
back of the shadows, of “the shadow/’ identified with the enemy,
barbarous, demoniac, “shapeless,” dangerous, etc., by Dawn or the
band of Dawns — noble goddesses, “women of the ary a,” aryapatnlbi
(7.6. 5)/ supdtnih (6.44.23). This is what the bonae matres , the uniuirae
ladies, also act out in the Matralia, against a slave woman who must
represent, in contrast with themselves, the wicked and base-born
element.

2. In this world freed from shadows. Dawn or the Dawns bring
in the Sun. This simple truth receives many expressions in the Vedic
hymns, but one of them, a singular one which the phenomenon itself
does not suggest, is probably the result of reflection by the priests.
Dawn is the sister-goddess par excellence. In the QgVeda the word
svdsr “sister” is applied only thirteen times to a divinity; in eleven
cases it refers to U$as or to a divinity called a sister of U$as; and it is
with Night, Ratrl, a divinity of the same type, that she forms the most
constant “sisterly couple.” In the eleven texts just cited, six concern
Usas as sister of Ratrl, or vice versa; there are five examples, in the
dual, of “the two sisters,” three times designating Usas and Ratri, and
twice Heaven and Earth. And this is not merely an elegant figure of
speech; the expression is a good definition of the relations between
these two persons. To the same extent that Dawn acts violently
against the demoniac Shades, she is respectful and devoted toward
Night, who, like herself, belongs to the grand scheme of the world,
this rtd or cosmic order, of which they are conjointly called “the
mothers” (1, 142.7; 5.5.6; 9.102.7). But it is another child of these
collaborating mothers who gives rise to the most characteristic
expressions : either, by a peculiar physiological process, they are the
two mothers of the Sun, or of Fire, sacrificial or otherwise, “their
common calf” (1.146.3; cf. 1.95.1; 96.5); or. Dawn receives the son,
the Sun or Fire, from her sister Night, and cares for him in her turn. 7

6. Despite the accent, this is the most probable meaning of this word, which is applied in
the R V once to the Dawns and once to the Waters.

7, The importance and even the authenticity of this mythical representation of the
Sun's two mothers have been thoughtlessly contested by J. Brough. It receives a striking
confirmation in the Wikanderian exegesis of the Mahabhdrata. The chief heroes of this
poem are presented as the sons of the chief gods; in character and behavior they reproduce
the essential characteristics of their fathers/ Thus Karna, the son of the Sun god, duplicates
three mythic traits of the Vedic Sun; (1) His hostile relations with the hero Arjuna.the son

THE LOST MYTHOLOGY

This second form of expression is especially useful here, as is shown by
( he following illustration (3.55.11-14):

1 1. The twin sisters [ yamid ] have put on different colors, of which one shines
while the other is black. The dark and the red are two sisters [ svasdrau ] . . *

13. Licking the other’s calf, she bellowed [anyasyd vatsdm rihatimimdya ] . . .

14. The multiform one dresses herself in beautiful colors, she holds herself
upright, licking a year-and-a-half-old calf.

Throughout these variants the governing idea is constant: Dawn
suckles (1.95.1 and 96.5) or licks (3.55.13) the child who either belongs to
herself and her sister Night in common, or to that sister alone. Thanks
10 this care, this child, the Sun (or in the liturgical speculations, the
1 ’ire of the offerings, and all Fire), which has emerged from the womb
of Night, arrives at the maturity of day. These mythic expressions,
which articulate the ideas of "mothers,” of “sister,” of “the sisters
child,” well express the function of the brief dawn: the appearance of
a sun or a fire, which was already fully formed before she intervened.

I ike the first, the second rite of the Matralia is completely clarified
by this confrontation with Indian ideology. The Romans, a realistic
people, simply cannot envisage any physiological prodigy. The
child does not have two mothers but, as in the variant which the
Indians did not prefer, a mother and an aunt. The Sun, son of the
Night, is taken in charge by her sister, the Dawn. 8

Finally it should be remarked that of all the divine figures of the
KyVeda Usas has the most highly developed kinship. 9 Sister of Night,
mother and aunt of the Sun (or of Fire), she is also in other contexts
i lie wife or lover of the Sun (or of Fire), and of him alone (for she is
not a courtesan), daughter of Heaven, and mother in general, both of
men (7.81.4) and of the gods (1.113.19). Thus the matrons who act
oiu in the Matralia the two aspects of "Mater” Matuta's function

• •I Indra, are those of the Sun and Indra. (2) Arjuna overthrows him when a wheel of his
■ 1 1. 1 not sinks into the earth, just as Indra detaches a wheel from the chariot of the Sun. (3)
1 ike t lie Sun, he has two successive mothers, his natural mother who abandons him on the
very night of his birth, and his adoptive mother whom he later acknowledges as his true
moi her; see ME 1 : 126-35.

h In it necessary to emphasize that this representation of Dawn does not presuppose the

• mm cnee at Rome of a parallel Night goddess? The Vedic Night goddess, the sister of
l ».iwn, hardly exists outside the myths of the latter.

o l„ Rcnou, Etudes vediques ct panineennes , 3, pt. 1, Les hymnes d VAurore du RgVeda
( • w) : H~9; the entire brief but substantial introduction (pp^ 1-12) should be read.

PRELIMINARY REMARKS

and who must therefore not only act like but "be” like her, are at the
same time sisters, mothers, and aunts, and fulfill a further condition
with respect to their husbands by being uniuirae. No other Roman
ritual carries the cumulative demands for so many familial relation-
ships.

The information provided by this comparison has far-reaching
consequences. The two actions of the bonae matres and the fa^nilial
conditions which they had to fulfill have not been explained to us by
the Romans themselves, and Plutarch, who tried to find Hellenizing
justifications for them, was not aware of this ancient meaning. Per-
haps it had been totally lost. And yet, behind the actions, at Rome as
in India, there must have been in primitive times ideas, a nuanced
and complex portrayal, negative and positive, rational but also dra-
matic, of the phenomenon of dawn, or rather of Dawn, the Dawns,
conceived as divine persons. Because, and we must insist on this
point, neither of the two Roman rites is directly inspired by the
phenomenon or acts it out objectively; both substitute an anthro-
pomorphic interpretation of it, with connections of kinship — a
noble sister-goddess of the night hostilely driving out her opposite
who has the marks of a slave woman, then affectionately fondling
her nephew the Sun; there is also, as in India, a multiplication of this
divine person in the group of “Dawns,” acting as a team. Thus there
formerly existed a Roman mythology of the dawn in the common
Greek or Vedic sense of the word, and we can read its essential features
as they are preserved in the ritual.

If the first of the two representations thus revealed — the Dawn
ladies driving out the wicked Shade — may have been formed in-
dependently in India and in Rome, the second — the Dawn ladies
fondling their sister’s son — is original and even unexpected. I know
no example of it outside of India and Rome; it is thus a priori probable
that the mythology of the dawn was inherited, in these two societies,
from their common past . 10

Finally this mythology is on an advanced level. It assumes an anal-
ysis less of the phases than of the effects of the dawn phenomenon,
and a subtle distinction between the wicked shade driven out by the
Dawns and the fecund Night whose fruit they gather. Here the ear-
liest Romans are restored to their proper intellectual dignity, to the

io. On the annual date of the festival of Mater Matuta, see below, pp. 337 - 39 -

THE LOST MYTHOLOGY 55

level of the Vedic bards, not, to be sure, for the poetic expression but
lor the conception. We are far from the stammerings, the impotendes,
and even from the “electrodynamic land of enchantment” to which
t he primitivists reduce them . 11

At what stage did Rome lose its mythology? One can only guess.
Personally I would not put the beginning of this process very far
back. In any case, it was certainly hastened and completed by the
invasion of Greek mythology, which was much richer and much more
prestigious. The fate of Matuta in the era of the interpretationes is
interesting. As no Greek goddess corresponded to her in her central
function (rosy-fingered ’Hus is only a literary figure), she was learn-
edly assimilated to Leukothea, by virtue of two analogous details in
t he ritual. Although in the minds of the Romans, she remained the
dawn goddess, the transferral of Leukothea's mythology to Matuta
caused the latter s andent mythology to be forgotten; on the other
hand, it created a connection between her and a divinity who in
primitive times had not concerned her in any respect: Portunus,
who was assimilated to Palaimon, and became her son.

There is another question which we should like to ask in a useful
manner but which, for lack of documentation, remains open. Had
i he other Latin peoples also lost their mythology? Had their divinities
become as emadated, as abstract, as those of Rome? A few rare
indications suggest that certain ones had been less radically so. The
I or tuna Primigenia of Praeneste, in the enigmatical conflict which
defines her, is portrayed as being at the same time primordial,
fn imigenia , and, contradictorily, the daughter of Jupiter, puer Jouis.
As this conflict seems to be fundamental (it recalls a Vedic theologem
i m the same level), we are tempted to admit that the connection with
| u pi ter is an ancient one . 12 At Praeneste, again, there exists an in-
separable pair of brothers whose typological kinship with Romulus
,md Remus is certain; nevertheless they are gods, and as gods they
miervene in the story of the founding . 13 This being so, we should

1 1 . H. Wagenvoort, Roman Dynamism , p. 79. Recently, passing from one extreme to the
m Ikt, the same author claimed to find in Rome the traces of an Indo-European conception
• ii Paradise, “Indo-European Paradise Motifs in Virgil’s IVth Eclogue,” Mnem. 4th ser., 15
(imma): 133-45. Greater precautions are needed in making Indo-European reconstructions.

ii. This is the subject matter of chap. 3 of DL. Primigenius does not mean “eldest”
(with respect to brothers), but “primordial.”

u. Below, p. 253-

5 6 PRELIMINARY REMARKS

perhaps not dismiss the account which, at Rome itself, makes the
twin founders the sons of Mars, for the sole reason that divine relation-
ships are foreign to the religious thought of the earliest Romans. Nor
need the other variant, which also has a parallel at Praeneste, be a
late invention. In this version the Roman twins are engendered by a
phallus which appears on the hearth. Greece ignores this theme, and
thus could not have furnished the story, but it is found in India, where
the cult of the hearth bears a strong resemblance to that at Rome.
Although it is not attested in Vedic literature, it occurs in many epic
traditions which may be ancient, even if they were written down at a
later date: Karttikeya, god of war, is born from the desire which
Agni, the personification of fire, conceives in the presence of noble
women; he enters the domestic hearth ( garhapatya ), and with flames
(. sikhabhih ) for organs, satisfies his lusts . 14 Once again, however, the
instances of this genre are too rare, and the religion of the other
cities of Latium are too poorly known to permit any judgment.

If it has been proved that the Rome of the eighth century knew
more mythology than the Rome of the third century, one must still
realize that for lack of poets, and in contrast with Greece, this mythol-
ogy was not literary or exposed to the temptations of literature, but
rather was restricted to useful functions and "glued” to ritual. In
this respect it was already deserving of the praises which Dionysius
of Halicarnassus gives to the theologia of his era. Jupiter O.M., when we
first meet him, is a serious and completely respectable gentleman,
and it is highly unlikely that the Romans or their ancestors attributed
to him the amatory adventures of his Hellenic homologue.

Finally, our peculiarly favorable situation with respect to the
Matralia is not usual. Except for Matuta and Angerona there is
almost no divinity whose ritual or figuration acts out or signifies a
picturesque mythical activity which we can interpret completely. In
the Lupercalia the obligatory laughter of two boys and the blood
rubbed on their foreheads remain unexplained, doubtless for lack of
parallels among other peoples, and also for lack of exact knowledge of
who the divinity of the festival was. The Nonae Caprotinae may rest

14. Romulus: Plut. Rom. 2.7. Servius Tullius: Dion. 4.2. 1-4, etc. Caeculus (Praeneste):
Serv. Aen. 7.678; A. Brelich, Vesta (1949), pp. 70, 97-98. See Buchheit, Vergil fiber die
Sendung Rams (1963), p. 95. n - 38o. Karttikeya: Mbh. 3.14291-92. Agni and the daughter of
Nila: MWt. 2. 1124-63-

THE LOST MYTITOLOGY 57

on a lost myth of Juno; we cannot even guess at it. In fact, the justi-
fications for a great many of the Roman festivals — at least for those
which have any — are of a different type. They do not concern the gods,
hut rather men, the great men of the past. The race of the Luperd
and their two teams supposedly arose from a circumstance in the
life of the twin founders, and the flagellation on the same day from a
calamity in Numas time. The fires lighted on the Parilia are said to
derive from an episode during the founding. A surely fictitious
episode of the Gallic siege, taken from the folklore of war strategems,
"explains” the cult of Jupiter Pistor. And so forth. But, for the greater
part of the time, the label "historical” refers only to the establish-
ment of the rites and does not explain their development. In this way
a great many little etiological legends were formed, some purely
human, others associating the gods with men. The greater number are
of late invention and without interest beyond their attestation of a
certain religious attitude. For example, the dickering between Jupiter
and Numa, mentioned above, or even the "justifying” myth of the
I arentalia on 23 December, which is a good specimen of the level of
I his literature. 15 One day during the reign of Ancus Marcius, when the
aedituus of the temple of Hercules was feeling bored, he proposed to
the god that they gamble together. The stakes: the loser would serve
t he winner a sumptuous feast and would procure a beautiful girl for
him. He rolled the dice, one hand for himself, the other for the god,
and having lost, discharged his obligation. He placed the promised
feast on the altar and locked into the temple the most celebrated of the
light women of that time, Acca Larentina (Larentia). A flame coming
out of the altar consumed the food, and the lady dreamed that the
god enjoyed her and promised her that the first man she should
1 nect on the following day would give her the customary little present.
In fact, when she left the temple in the morning, she met, according to
Mime a young man, according to others a graybeard, but in any
< ase a very rich man who loved her, married her, and died leaving her
an enormous inheritance. She in turn bequeathed this property,
notably the real estate, to the Roman people, who in acknowledg-
ment sacrifice on her tomb at Velabrum every 23 December, with
important priests in attendance. There is of course nothing of value
in this story, unless it is the suggestion that the recipient of the Laren-
1 ■>- Plut. Rom . 5.1-10, and Q.R. 35; Macr. 1. 10.13-16. On aedituus , see Latte, p. 410.

58 PRELIMINARY REMARKS

talia had some ideological connection with wealth and pleasure. But
her mythology as such has disappeared.

This kind of almost completely demythologized religion, surviving
only in rites whose mythological and even theological justifications
have been forgotten, is seldom found in other parts of the Indo-
European world. The most remarkable case is that of the Indians of
the former Kafiristan in the Hindu Kush. This region becarn^ Nuris-
tan at the end of the nineteenth century, when the Afghans imposed
on it the "Tight” of Islam, nur, with the violence appropriate to this
kind of attention. Living in their high valleys, the w Kafirs,” an in-
telligent and handsome people, had preserved until then an interesting
religion, the features of which recall Vedism and which was observed
in extremis by an English traveler, Sir George Scott Robertson, whose
book is one of the masterpieces of ethnography. 16 Robertson devoted
two admirable chapters (23 and 24) to this religion, but he begins by
making excuses which an observer of the Roman religion of the
final centuries might have made if he substituted Hellenism for
Islam and Athens or Rhodes for Chitral.

It must be remembered that the Bashgul K&firs are no longer an isolated
community, in the strict sense of the word. They frequently visit Chitral
and have dealings with other Musalman peoples as well. Many of their
relatives have embraced Islam without abandoning the ties of relationship.
One of the results of this free intercourse with Musalmans is that the Bashgul
Ksifirs at the present day are very apt to mix their own religious traditions
with those of their Musalman neighbors. This greatly confuses matters, and it
is hopeless for me to try to write anything final, or even moderately compre-
hensive, concerning the religion of Kafiristan; a modest record of what I
actually saw and actually heard is all that can be attempted. Possibly a better
acquaintance with the Bashgul language might have made many things clear
to me which now remain dark, and perhaps had my interpreters been better
the same result might have followed; but it appears to me that the chief
reason why I discovered so little about the K&fir faith is because the Kafirs
themselves know so little on the subject. It would seem that in Kafiristan the
forms of religion remain, while the philosophy which those forms were
originally intended to symbolise is altogether forgotten. This is not, perhaps,
surprising in a country in which there are no records of any kind, and every-
thing depends on oral tradition.

16. The Kdfirs of the Hindu-Kush (1896). Robertson’s visit took place in 1889-90. The lines
cited here are on pp. 378-79*

THE LOST MYTHOLOGY

The Bashgul Kafirs, or at any rate the younger portion of the community,
are inclined to be somewhat sceptical They are superstitious, of course, but
^crcd ceremonies are frequently burlesqued or scoffed at when two or three
waggish young men get together. Gish (the god of war and of warriors) is the
really popular god of the Bashgul youth. In their worship of him there is a
great sincerity. A young Kafir once asked me if we English did not prefer Gish
to Imra (the Creator), as he himself did, and many Kafirs have expressed their
disappointment on learning that Franks knew nothing of Gish.

The older people are devout in their respect for all the gods, but Bashgul
Kafirs seem ready to abandon their religion at any time without much regret.
They leave it, as they return to it, chiefly from motives of material advantage,
and rarely appear to trouble themselves about religious convictions.

These lines would be an excellent preface to the Fasti of Ovid.

FROM MYTHOLOGY
TO HISTORY

By means of the substantial figure of Mater Matuta we have been
brought back to the fundamental fact, already mentioned several
times: the Indo-European heritage. In its organization of the state
and in its religion during the earliest stages Rome reveals itself as the
continuator of an Indo-European tradition. To what point does this
heritage extend? Are other, non-Indo-European elements discernible
before the Etruscan influence, before the Greek invasion? In what
proportion were they associated and intermingled with one another?
In short, we arc confronting the problem of the substratum. How is
it posed today?

A half-century ago Andre Piganiol proposed and developed a simple
solution . 1 He admitted the authenticity of the <f synoecism” which the
annalistic tradition assigns to the beginnings. According to his solu-
tion, Rome was formed by the union of Latins and Sabines, the former
being Indo-Europeans and the latter Mediterraneans, and the con-
siderable contribution of each of the components could be recognized
in the organization, in law, and in religion. Especially with respect to
religion, the Indo-Europeans were responsible for the burials by
cremation, the Sabines for the interments, which are juxtaposed and
sometimes overlap in the sepolcreto of the Forum. The Indo-Europeans
brought into Italy the altar bearing a lighted fire, the cult of the
male fire, that of the sun, that of the bird, and a repugnance toward
human sacrifices; the Sabines used stones which they rubbed with
blood as altars, ascribed the patronage of fire to a goddess, offered
worship to the moon and to the serpent, and immolated human
victims. This construction was and could be only arbitrary. Apart

I. Ettdi 5nr les origines de Rome (1916). Discussed in Dumezil, NR, chap. 3 (" Latins et
Sabins: histoire etuaythe 1 ").

FROM MYTHOLOGY TO HISTORY 6l

from the fact that the Sabines, who were themselves Indo-Europeans,
could not play the role of “Mediterraneans” which was entrusted to
them, there was then no means of taking an objective view either of
Indo-European civilization or of Mediterranean civilization. The
author was thus free, too free, in dividing between these peoples the
two terms of many antithetical pairs of beliefs, practices, or institu-
tions which he claimed to observe in historical Rome. In later works,
some of the most obviously debatable statements have been modified
by the author, but he seems to have maintained the principle of the
double origin of Roman civilization and to have confirmed the Sa-
bines, against all probability, in their role of Mediterraneans. To the
best of my knowledge, he has not disavowed the summary and
ti' priori method which governed these dichotomies and the distribu-
tion of their terms between the two components . 2

For the past thirty years the idea of Indo-European civilization has
progressed appreciably by the only admissible procedure, compari-
son. The comparative study of the most ancient documents from
India, Iran, Rome, Scandinavia, and Ireland has allowed us to give it
.» content and to recognize a great number of facts about civilization,
,ind especially religion, which were common to these diverse societies
or at least to several of them. Many of these agreements, such as the
one which has just been summed up concerning Dawn, are so singu-
lar that they cannot be interpreted otherwise than as the heritage of
Indo-European conceptions. Several also are interdependent, comple-
menting and articulating one another in such a way that it is not an
inorganic dusting of more or less considerable concordances which we
find but entire structures of representations. This is not true of the
notion of a “Mediterranean civilization .” 3 Despite many worthy

i , In 1949, however, in a public course, the author sketched out another analysis of the
lacis of Roman religion which it does not seem easy to reconcile with the Latin-Sabine
dichotomy and which we should like to see developed. All that I know about it is what is
printed in the Anmaire du College de France, 1950, p. 200: “The course has been devoted
.1 1 most entirely to the problem of the origins. The professor has sought to link the study
oj The religious phenomenon closely with that of economic and social life, and to discover
m the series of festivals the rhythm of the life of the nomadic shepherds, the sedentary
1. 1 miers, and even, more distantly, the hunters. He has had to take up once more the
problem of the Indo-European religion.”

\. Despite my disagreement on many details of method, I pay my respects to the labors,
notably, of Mme L. Zambotti, II Mediterraneao , VEuropa, lltalia durante la preistoria
(1054); of U. Pestalozza, Pagine di religione mediterranea , vols. 1-2 (1942-45), and Religione
mediierranea (1951). For a wealth of information and a debatable method, M. Marconi,
l< ijlcssi mediterranei tteUapiu antica religione locale (1939).

PRELIMINARY REMARKS

efforts, this notion remains confused and continues to thrive on the
arbitrary. Here too, the comparison of several related areas would be
the only possible method of exploration, but there is not the point of
departure, the support, which linguistic kinship provides for the
Indo-Europeans. The general unity or the great partial unities that are
presumed in the Mediterranean basin, regardless of the unknown
languages and the indeterminable ethnic relationships and migra-
tions, are necessarily uncertain. Moreover, the data which are used
are almost exclusively archaeological, a fact which allows, as we know,
the most generous interpretations.

Fortunately, the question of the substratum no longer has the
importance which the Essai of 1916 attributed to it. In contrast with the
Greeks who overran the Minoan world, the diverse bands of Indo-
Europeans who descended on Italy certainly did not have to confront
great civilizations. Those among them who occupied the site of Rome
do not even seem to have been preceded by a dense and stable
settlement, and traditions like the one dealing with Cacus suggest
that some of the indigens encamped on the banks of the Tiber were
as simply and summarily ousted as the Tasmanians were to be, in
the Antipodes, by the traders coming from Europe. 4 It will not be
concluded that everything in primitive Rome was a heritage from
Indo-European ancestors: much may have been created on the spot
to fill new needs; much also, as happened frequently in the following
centuries, may have been borrowed from the other peoples of the
peninsula, who were themselves for the most part heirs to the Indo-
European past. But it is out of the question that there could have been
at Rome a mixture in equal parts of Indo-Europeans and pre-Indo-
Europeans.

If this question of the substratum thus loses its importance, it is
replaced by another: the question of the homogeneity or the duality
of the Indo-European founders of Rome. Was Rome born out of the
development, or the conquests, of one group or several closely asso-
ciated groups of Latins, or was it produced, as some claim they can
infer from the annalistic tradition, by a fusion of this group or groups
with a group of Sabines? As can be seen, the terms are appreciably the
same as in PiganioTs problem, but they have a different coloration,

4. The traditions about the Aborigines seem to be purely legendary and scholarly.

FROM MYTHOLOGY TO HISTORY

since the two component elements are now both Indo-European.
The stakes are no less important: they are the meaning of the god
Quirinus, and consequently the meaning of Rome's oldest known
theological structure, the triad formed by Jupiter, Mars, and Quirinus.

The greater number of modern historians, although with growing
hesitation, continue to incline toward duality, toward synoecism. 5
Considering the actual state of the dossier, this is an act of faith which
is not supported by archaeology or by toponymy or even by the
annalistic tradition as it is clarified by external criteria.

A number of authors assume, as Piganiol did in his Essai and as F.
von Dahn did in 1924 in the first volume of his Italische Grcfberktinde ,
that the duality of burial practices, cremation and inhumation, is a
serious presumption in favor of the duality of population. It is not.
In many countries and many eras, the two ways of treating the corpse
have coexisted without there being any difference of race or language
between those who practiced them, and even without any difference
in the manner of conceiving the beyond, the life after death. Not to
leave the Indo-European world, the Vedic Indians practiced them
concurrently. Cremation is dominant in their texts, but the four-
teenth strophe of the funeral hymn R V 10.15 mentions conjointly the
ancestors who were consumed by fire and those who were not, while
strophes 10.18.10-13 certainly refer to inhumation; however, all the
dead go to Yama. In Scandinavia, where civilization retained a high
degree of homogeneity and where the movements of peoples were re-
stricted for a long time, the explanation of the instability of funerary
practices by ethnic differences or changes has been abandoned. Of the
first two centuries of the Christian era, a period when the influence of
Rome was felt, J. de Vries comments:

. . . the variety of funerary usages is particularly great. While in Gotland
tombs with interred bodies and offerings of weapons are predominant, in
Norway and Sweden cinerary burials are in the majority, even though in-
humations are found there too. In these two countries offerings of weapons are
the rule, while they are completely lacking in Denmark. Undoubtedly we
must admit that these differences are due to local and historical circumstances.
In any case, they are not qualified primarily by divergences in representations
of life beyond the grave, but rather by the variable currents of commerce, and
consequently of culture, acting on Scandinavia . 6

5. This, it seems, is also Latte’s position, p. 113 and n. 2.

6. Altgermanische Religionsgeschichte I* (195 6), §109, p. 148.

64 PRELIMINARY REMARKS

A little later, when the entire Germanic world was unsettled by the
Volkerwanderung, it is certainly possible that the new provincial
distribution of practices can be explained in part by the displacement
of tribes, but, as the same author notes, “in the course of this period
the mixing of the different kinds of tombs is pushed to the point
where one finds incinerated and interred remains in the same family
burial plot .” 7 Viewing the sepolcreto of the Forum and the sparse
necropolises of the Roman hills, we must not forget these examples . 8
On this subject Raymond Bloch remarked, before obviously senti-
mental reasons led him to support the opposite theory, “Latium
occupies a marginal position, peripheral with respect to the zones
from which emerge, respectively, the typical Villanovan and the
sub-Apennine iron civilization, namely, southern maritime Etruria
and the Osco-Umbrian territory. These two neighboring zones may
explain the interaction of cultural influences of diverse origins on
Latin territory and particularly on the site of Rome .” 9

Ethnic duality would seem probable if the traces of material civili-
zation, notably ceramics and ornaments, were appreciably different
on the Palatine, the Esquiline, and the Quirinal. This is not the case.
F. Villard, who has meticulously examined the ceramics on the site
of Rome between the eighth and the fifth centuries, has not found
any difference, and with respect to the second half of the eighth
century — the beginnings — he believes accordingly in the homo-
geneity of the population . 10

The remaining arguments for duality are feeble. That the northern
heights of Rome are called colles ( collis Quirinalis , collis Viminalis) as
opposed to the monies which lie beyond the Forum — the Palatine, the
Caelian, the Cispian, the Oppian, the Fagutaline— is certainly an
important difference which may have an explanation in history,
bearing witness for example to a progressive extension. But collis
is a Latin word with the same claims as mons, and nothing allows us
to say that the colies were originally Sabine and the montes Roman.

7. Ibid., §114, p. 152.

8. See the discussion of H. Mulier-Karpe, Vom Anfang Roms , KM, supplement 5 (1959),
PP- 36-39. By contrast E. Gjerstad, Gnomon 33 (1961): 378-82, takes his position in the line
of F. von Duhn.

9. Les origines de Rome (1959), p. 86.

10. "C£r antiques des premiers siccles deRome, VIII e -V* siedes ” (unpublished), according
to the analytic report (by A. Piganiol) in CRAI (1950), p. 292.

FROM MYTHOLOGY TO HISTORY 65

These variations are instructive concerning chronology but not
concerning the nature or the nationality of the population.

An attempt has also been made to deduce ethnic duality from
geography or, to use a recent term, geopolitics. We are told that the
then marshy valley of the Forum and Suburra was a veritable
frontier: on one side the heights of the Palatine, the Caelian, and the
Hsquiline formed as it were a projection of Latium toward the Tiber;
on the other side the Quirinal, extended by the Capitol and
doubled by the Viminal, played the same role, at least with respect
to those of the Sabines who also came down toward the Tiber via the
salt route. Was it not natural that each of these groups should seize
the advantage and occupy, on an important commercial site, the
positions which were offered to it? There is no need to point out the
subjective nature of these speculations. If one agrees to play this
game, it can also be objected that the most “interesting” northern
hill for the Sabines to occupy was not the Quirinal, but the Capitol,
near the Tiber and facing the Palatine; it might also be remarked that
it was “natural” for the Latin possessors of the monies to occupy
forward positions on the Quirinal from which an enemy might
threaten the Esquiline. But what good purpose is served by this
display of ingenuity? Geopolitics may help to account a posteriori for
known history, but it does not allow us to reconstruct unknown
history. 11

In fact, if one were to judge solely from the evidence of the excava-
tions or from a map of the hills, one could not possibly assume an
ethnic duality, a juxtaposition and then an association of Sabines and
Latins. The annalistic tradition, however, under conditions which
differ radically from those that have been deduced from the archae-
ological evidence, tells a dualistic story. Whatever one may pretend,
it is chapters 8-13 of Livy s first book and the parallel texts which
continue to form the principal argument. The grounds, the tombs,
and the ceramics have been subjected to interrogation, and sometimes
to entreaty, to supply evidence in the light of the traditional story of
t he origins. Thus, after loyally acknowledging that archaeology does
not prove the “concrete and actual coming of some tribes, each con-
iributing a different civilization,” Raymond Bloch returns to the

11. Cf. JMQ 4: 182.

PRELIMINARY REMARKS

vulgate: “ However, the agreement between the massive data of the
annalistic tradition and the elements of information provided by the
excavations seems sufficient for one to admit, on the whole, the idea
that two diverse populations, Latin and Sabine, were present in the
origins of Rome.” 12 Similarly, after making a bold interpretation of
the topography, Jean Bayet concludes, “Thus the earliest annalistic
tradition takes on verisimilitude: from a double settlement, Lati^ and
Sabine, of the site; from a double religious contribution, ascribed to the
first kings, the Alban Romulus and Numa of Cures, but under the
less schematic aspect of progressive relations (of war and peace)
between the originally distinct inhabitants.” 13 We have seen the
uncertainty in which we are left by topography and the “elements of
information provided by excavations.” What then are “the massive
data of the annalistic tradition,” “the earliest annalistic tradition,”
and what is their meaning?

The story of Rome’s first war is very well fabricated, but it is plainly
a fabrication. In the clearly revealed characters and the advantages
held by each side; in the sequence of well-balanced battle episodes,
none of which is decisive and which bring into play, one after another,
these characters and these advantages; in the unforeseen yet basically
logical development which turns a desperate war into something
better than an alliance, an intimate fusion; beneath the armed ma-
neuvers and the human passions we see a game of a different kind
being unfolded, described, and demonstrated, a rigorous game of
concepts. Understood in this way, the story of the formation of the
full Roman society is exactly parallel to those accounts, not “his-
torical” but mythical, which are known to other Indo-European
peoples. These accounts tell how the full society of the gods was
formed, starting with two groups originally juxtaposed, then opposed
in a war with alternating victories and defeats, and finally united in a
true fusion. I reproduce here the comparative analysis of the Roman
and Scandinavian accounts which I published in 1949 in Uheritage
itido-europeen d Rome , summing up earlier works :

I. Here in the beginning, before the war, is the description of the two op-
posing sides:

12. Lesoriginesde Rome t p.S 6 .

1 3. Histoire politique et psychologique de la religion romaine (1957), p. 23.

FROM MYTHOLOGY TO HISTORY

1. On one side, Romulus. He is the son of Mars and the proteg£ of Jupiter.

I le has just founded the city in ritual form, having received the auspices and
marked out the sacred furrow. He and his companions are magnificent youths,
sirong and brave. This side has two trump cards: it has the great gods with it
.md partially in it, and it is full of warlike qualities. On the other hand, it
has gross deficiencies in terms of wealth and fecundity: it is poor, and it is
without women.

2. On the other side, Titus Tatius with his wealthy Sabines. To be sure,
they are neither cowardly nor irreligious — quite the contrary — but at this
point in history they are defined as rich. Moreover, they possess the women
that Romulus and his companions need.

Before they confront each other, before they even dream of confronting
each other, the two sides are thus complementary. And it is because they are
< omplementary that Romulus, realizing that his incomplete society cannot
survive, has “ the Sabine women ” carried off in the course of the rustic festival
nl Consus. He acts in this way both to obtain the women and to oblige the
rich Sabines, despite their repugnance, to enter into relations with his savage
band.

All the authors agree in stressing and making explicit this conceptual,
functional motive of the earliest events. Reread in Livy 1.9.2-4, the instruc-
tions which Romulus gives to his ambassadors when, before resorting to
violence, he sends them to the surrounding cities. They are charged to tell
1 heir future fathers-in-law :

“ Cities , . . . as well as all other things , take their rise from the lowliest beginnings .
\s titne goes on , those which are aided by their own valor and by the favour of
the gods achieve great power and renown . They said they were well assured that Rome’s
ongin had been blessed with the favour of the gods , and that valor would not be
inking; their neighbours should not be reluctant to mingle their stock and their
blood with the Romans , who were as truly men as they were ”

I )/ and uirtus , the gods and courage or manly energy, define very well the
bases of the first two functions; opes , resources, power consisting of property,
ol 1 he means of action, and also the means of fertility and propagation, here
designated by sanguis ac genus , characterize the third equally well. Di, mean-
ing the divine ancestry of the two brothers and the promise given by the
auspices on the site of the future Rome, constitutes the double supernatural
dement which they bring as a dowry; as for uirtus 9 they have not yet tested it
m actual combat, but they feel it within themselves. Opes is the only factor
which they do not yet have, either potentially or actually, and which is not
ascribed to their nature. When they shall have acquired it and intermingled
w 11I1 it sanguinem , the synthesis of the three principles which were originally

PRELIMINARY REMARKS

distributed between the two neighboring peoples will assure Rome of its
place in history, nomen . And in fact the synthesis will have this result* Another
historian, Florus, summing up the war very schematically (i.i), writes that
after the reconciliation the Sabines moved to Rome and cum generis sins
audios opes pro dote sociant , sharing their hereditary riches, like a dowry, with
their sons-in-law*

In the third book of his Fasti (lines 178-99) Ovid provides the same con-
ceptual substructure to the event, but in dramatic form. It is the god^ Mars
himself who tells how he inspired his son Romulus with the idea of carrying
off the Sabine women :

“ Wealthy neighbours scorned to take poor men for their sons-in-law; hardly did
they believe that I myself was the author of the breed . . . . J chafed and said , ' Thy
father s temper , Romulus , I have bestowed an thee . A truce to prayers! What thou
seekest , arms will give”*

Here once more Romulus’s two trumps are first his divine birth, with a
god as auctor sanguinis, and second, thanks to a direct inspiration from this
god, a warlike temperament, patriam mentem , and arms, arma. His opponents
are rich men, uicinia dines , who scorn his inopia.

Dionysius of Halicarnassus (2.30.2 and 37.2), wordy as always, and following
a slightly different tradition (involving not two, but three races, among whom
the three trumps are distributed), still expresses the same fundamental
structure. Sounded out by Romulus for matrimonial alliances, the Latin
cities refuse to join these newcomers “ who neither are powerful by reason of
their wealth nor have performed any brilliant exploit.” For Romulus, thus
reduced to his quality as son of a god and to the promises of Jupiter, there
remains nothing to do but to rely on professional soldiers, which he does,
summoning among other reinforcements Lucumo of Solonium, “a man of
action and reputation for military achievements.”

Such is the structure of the entire plot: the need, the temptation, the in-
tention, and the action of Romulus all have as their goal the formation of a
complete society by imposing on the “wealthy” the necessity of associating
with the “brave” and the “godlike.”

II. The war itself falls into two episodes. In each, one of the two sides is
almost victorious, but each time the original situation is restored, and a
final decision is postponed.

i. First there is the episode of Tarpeia. It is told in different ways, sometimes
(these are the most beautiful forms : Propertius . . .) with the passion of love
as the motive ; but it seems to have its purest form in the version adopted by
Livy (1.11.5-9). Titus Tatius, the head of the wealthy Sabines, using as a

FROM MYTHOLOGY TO HISTORY

bribe the gold, the bracelets, and the jewels which sparkle on the arms of his
men, seduces the daughter of the Roman who is entrusted with guarding
die essential position of the Capitoline. Treacherously admitted into this
dominant fortress, the Sabines seem about to be victorious. 14

2. In fact, they almost are, when the second episode occurs. This time it is
Komulus who seizes the advantage (Liv. 1. 12.1-9). In the course of the battle
in the valley of the Forum between Romulus's companions, who have been
driven back to the Palatine, and Tatius's Sabines, who control the Capitoline,

I lie former yield and fall back in disorder. Then Romulus raises his sword to-

ward heaven and says, " O Jupiter, it was thy omen that directed me when I
laid here on the Palatine the first foundations of my city But do thou . . .

I I diver the Romans from their terror, and stay. I here vow to thee, O Jupiter,
1 lie Stayer, a temple, to be a memorial to our descendants from the City saved
I >y thy present help/'

“ Having uttered this prayer he exclaimed , as if he had perceived that it was heard ,

' / lere t Romans , Jupiter Optimus Maximus commands us to stand and renew the
fight V The Romans did stand , . . . and . . .fired by the reckless daring of their king ,
drove the Sabines before them.”

Thus Romulus counters the criminal bribery ( scelere emptum) of Titus
Tatius with an appeal to the sovereign Jupiter, the greatest god whose aus-
pices have guaranteed the grandeur of Rome. And from this god he obtains an
immediate mystical or magical intervention, which against every expectation
reverses the morale of the two armies and changes the fortunes of battle.

We see the meaning of these two episodes and the way in which they are
applied constitutively to the descriptions of the two sides as they were first
presented: the Romans and the Sabines, Romulus and Titus Tatius, engage in
kittle, and on equal terms, and it is neither courage nor strategic skill which
distinguishes one from the other. But each of them, the chief of the wealthy
Sabines on one side and the demigod Romulus on the other, has his own way of
1 ni (‘evening in the battle and causing victory to incline toward himself. The
wealthy man with his riches has recourse to gold, to the shameful trick of
1 nrriiption, not by means of money at that time but by means of jewels— the
Imid of corruption most effective with a woman; the demigod obtains from
1 he all-powerful Jupiter the gratuitous miracle which changes a defeat into
vu lory. To understand the logical structure of this entire arrangement, it is
only necessary to state the impossibility of imagining that the roles are
irversed, that it is Romulus who resorts to bribery and Titus Tatius who

1 I- On Tarpeia, see my essay at the end of the book bearing this name (1947) and my
imir, RUL 38 (i960): 98-99. The theme is definitely borrowed from Greece, A. H. Krappe,
/c/iM 78(1929): 249-67.

PRELIMINARY REMARKS

obtains the miracle from Jupiter: this would have no meaning. Titus Tatius
and Romulus act not only in accordance with their characters, but in
accordance with the functions which they represent.

HI. How does the war end? No military decision has occurred. The demigod
has neutralized the wealthy man, the miracle of the celestial god has balanced
the power of gold, and the struggle threatens to go on forever. Then un-
expectedly the reconciliation takes place; the women cast themselves between
their fathers and their ravishers. And everything turns out so well that the
Sabines decide to merge with Romulus’s companions, bringing to them as a
dowry, as Florus says, auitas opes. The two kings become colleagues, and each
institutes a cult: Romulus to Jupiter alone, and Titus Tatius to a whole series
of gods connected with fecundity and with the soil, among whom Quirinus
figures. Never again, either under this double reign or later, will we hear talk
of dissension between the Sabine element and the Latin, Alban, “Romulean”
element of Rome. The society is complete. A chemist would say that the
valences of the diverse elements have saturated each other reciprocally. To
use Livy’s words again, Romulus’s group, which in the beginning had deos et
uirtutem on its side, has gained what it was lacking, opes , as well as the Sabine
women, the pledges of national fecundity.

We see that from one end to the other the logical connection, the significa-
tive intention, and the necessity of the episodes are clear. Everything is
oriented toward a single meaning, everything states and presents a single
lesson; it is the history, in three stages, of the formation of a complete city,
starting with its presumed preexistent and originally separate functional
components. First stage: the presentation of these separate incomplete
components, of which at least one, the superior component, is unfulfilled
and not viable as such; second stage: the war, in which each component
expresses its genius in a characteristic episode (gold on the one side, grand
magic on the other, dominating the combat proper); third stage: an associa-
tion of these components, unforeseen but nonetheless firm and definitive, in
a unified society. And the history of Rome begins.

Where did the Romans get this schema from? In principle, one might be-
lieve that they got it from nowhere and from nobody, that it is the peculiar
product of their genius, that they invented it, in the full meaning of the word.
But it is here that a comparison with the traditions of other Indo-European
peoples furnishes a light which Latin philology by itself cannot supply and
resolves an uncertainty from which literary criticism cannot escape by itself.
In fact, other Indo-European peoples also use an articulated history to explain
the formation of a complete society in accordance with the system of the
three functions, starting with originally disparate elements. I shall confine

FROM MYTHOLOGY TO HISTORY

7 1

myself to reproducing briefly the Scandinavian version of this tradition, the
account of the war, and then of the reconciliation of the dEsir and Vanir.

We are not concerned here with an ordinary human society, but a divine
m K'iety, the difference being further heightened by the fact that in at least
one of the usable texts the gods composing this divine society turn out to be
i he ancestors of a human, Scandinavian society, and that we pass insensibly
from one to the other. The story is known to us from two texts of the Ice-
landic scholar Snorri Sturluson and from four strophes of a fine Eddaic poem,
i he Voluspd , “The Prophecy of the Seeress.” As we might expect, hyper-
« t iiicism has attempted to deny all validity to these two testimonies. Eugen
Mogk tried to make Snorri a kind of forger, from whose works one can keep
m >i hing but what one knows from other sources ; and on the basis of extremely
unreliable arguments he claimed to have demonstrated that the four strophes
<il the Voluspd are irrelevant to the whole matter which here concerns us.
A double discussion, which there can be no question of reproducing or even
of summing up here, has proved the error of the argument based on the
» ondemnation of Snorri and the dismissal of strophes 21-24 of the Voluspd.

I Icre, in three stages, is the sequence of events.

I. The Scandinavians recognize two well-characterized tribes of gods, the
Tsir and the Vanir. The JEsir are the gods who surround Odinn and porr
f Asapdrr,” as he is sometimes called). Odinn in particular, their head, is a
combination of god, king, and magician, the patron of earthly chieftains and
sorcerers, the possessor of magical runes and generally of powers which allow
him immediate action in all his domains; p6rr, the god armed with a hammer,
is 1 he great heavenly battler, the giant-killer, whose most famous actions are
involved with punitive expeditions, and whom one calls on in order to win
in single combat. The Vanir on the other hand are the gods of fecundity,
wealth, and pleasure; myths and cults indicative of this quality grew up around
1 lie three principal gods of the Vanir — Njdrdr (whom Tacitus describes in his
i in mania as the goddess Nerthus), Freyr, and Freyja.

Snorri ( Ynglingasaga 1-2), who anthropomorphizes them to the highest
degree, localizes the JEsir and the Vanir, as neighbors but completely separ-
ate, in the region of the lower “Tanais,” near the Black Sea. One group in-
li.ihits Asaland or Asaheimr, with Asgardr as their castle-capital; the other
1 1 1 lubits Vanaland or Vanaheimr.

II. Second stage (Snorri, Yngl. 4, beginning; Vdluspd 21-24). The iEsir
mack the Vanir, and there ensues, as the poem says, “war for the first time
m i he world.” “6dinn,” says Snorri, “marched with his army against the
Vanir; but the latter resisted and defended their country; now one side, now
1 hr other, was victorious; each devastated the other’s country, and they
inflicted mutual losses on each other.”

PRELIMINARY REMARKS

From the precipitate, allusive poem we know the two episodes — the only
two — of the war:

1. A sorceress named Gullveig ‘"Frenzy (or Power) of Gold,” apparently one
of the Vanir or sent by them, comes to the Aisir ; the latter bum her and then
burn her again in OQinn's hall, but do not succeed in killing her altogether;
she continues to live as a witch; in particular, she “is always the delight of
wicked women.” **

2 . Odinn, the great magician-god, chief of the IE sir, hurls his spear at the
enemy, making for the first time the magical gesture which several of the
texts later attribute to human chieftains, and the intention of which they
specify: as is said in a comparable case, the Eyrbyggjasaga ( 44 * 13 ). it is a matter
of “gaining heill , luck, by magic”; and in the Styrbjarnar pdttr Svtakappa
(chap. 2 = Fornmanna Sogur 5 : 250), it is 6dinn himself who gives King Eric
of Sweden a canestalk and tells him to hurl it over the enemy army while
pronouncing the words, “6dinn possesses you all!” Eric follows the god's
advice: in mid-air the stalk turns into a spear, and the enemies flee, seized by
a panic fear. It is the prototype of this gesture which 65inn makes, a gesture
which should assure him of victory. Nevertheless, it does not succeed, since the
same strophe later describes the breaking of the lEsir’s rampart by the Vanir.

III. Worn out by this costly alternation of half-victories, the ALsir and the
Vanir make peace. An unforeseen peace, as complete as the war was desper-
ate; a peace whereby, at first as hostages, then as equals or “nationals,” the

15. Quite recently J. de Vries, who approved on the whole of my analysis, has proposed
a new exegesis of strophes 21-22 of the Voluspa, which seems to me too critical and at the
same time too free (ANF 77 [1962]: 42-47). Moreover, if he were right, we would still have
what the two allusions of these lines tell of the Vanir people on the one side, and 6dinn on
the other, each using his characteristic means against the opposing side: the Vanir’s seidr and
6dinn's spear. I think, however, that these two strophes need to be clarified and under-
stood with the help of the most continuous forms of the myth: (1) Snorri assures us that
the two sides achieve alternating advantages, without any decisive result; (2) the plagiarism
of the myth by Saxo Grammaticus (1.7.1), with his account of the golden statue sent to
Othinus and of the corruption caused by this gold in the heart of Othinus’s wife, is con-
firmation that we should give the name Gullveig its full meaning, and that line 22.4 does
not allude to the “incest” of the Vanir, but to the corruption of the Ase woman by her
desire for gold. I refer the reader to chapter 7 of my Saga de Hadingus , du mythe au roman
(i953)» especially pp. 105-11 (cf. Les dieuxdes Germains (1959], chap. 1). I still prefer what J. de
Vries has said about the war of the ALsir and the Vanir in Altgermanische Religionsgeschichte
II 2 (1957), pp. 208-14; cf. W. Betz, Die altgermanische Religion , in W. Stammler, Deutsche
Philologie im Aufriss 2 , cols. 1557-58 and passim. I cannot discuss here Heino Gehrts, “Die
Gulveig-mythe der VoluspS,” Zeitschrift fur deutsche Philologie 88 (1969) : 321-78. 1 only re-
mark that there is no internal contradiction (p. 359) in my treatment of Odinn's spear-
thrust. Of course, the myth is etiological, but what it justifies is not this particular gesture of
6dinn, but the constitution of a complete society; for this reason 6&inn’s gesture, which
normally could not miss its intended effect, victory, must fail in this particular case: just as
Romulus, who in all other wars could not miss victory, must be unable to win the Sabine
war, the end of which must be a compromise, a reconciliation.

FROM MYTHOLOGY TO HISTORY

principal Vanir, the gods Njordr and Freyr and the goddess Freyja, by the
fecundity and wealth which they represent, come to complete the society of
i he gods of 6dmn. So well do they complete this society that when “King”
( )5inn dies (for in the Ynglingasaga the gods are a kind of supermen who die
like ordinary mortals), it is Njordr, and after him Freyr, who become kings of
the /Esir. Never again, in any circumstance, is there the shadow of a conflict
between the JEsir and the Vanir, and the word ‘TEsir,” except when the
contrary is specified, designates Njordr, Freyr, and Freyja as well as Odinn and
|>6rr.

There is no need to stress the exact parallelism not only of the ideological
values which provide the point of departure but also of the intrigues from the
beginning to the end, including the two episodes which describe the war
between the rich gods and the magician gods. It seems hardly imaginable that
chance should have twice created this vast structure, especially in view of the
fact that other Indo-European peoples have homologous accounts. The
simplest and humblest explanation is to admit that the Romans, as well as the
Scandinavians, received this scenario from a common earlier tradition and that
i hey simply modernized its details, adapting them to their own “geography,”
** history,” and customs and introducing the names of countries, peoples, and
heroes suggested by actuality. 16

This explanation accommodates itself to Mommsen's. It gives an
account of themes, while Mommsen's is concerned with names . 17
I ( the preexisting, Indo-European account of the formation of a com-
plete society appears here exemplified in the ethnic framework of the
Latins, the Sabines, and eventually the Etruscans, it was the later
demographic movement of Rome which suggested it; in the fifth
and fourth centuries, as we know from definite clues, an afflux of
Sabine population actually did settle in the already established dty
and on the hills to the north. If the final treaty establishing synoedsm
was framed in the terms and style of which we read, it is in the image
of the collatio ciuitatis which in 290 b.c. actually did somehow merge
1 he twosodeties . 18

16. Hir . pp. 127-42 ; see above, p. 7, n. 3, and below, p. 73, n. 18.

17. Above, pp. 21-23. If it should be possible to demonstrate some day, archaeologically,
rfi.it there was an ethnic duality in the origins of Rome, the account given by the annalistic
1 1 .idition would still remain entirely "prefabricated”; the historicized myth would merely
he* superimposed on history.

i H. See now my expose of the second part (“Naissance d'un peuple”) of ME 1 discussing
especially J. Poucet, Recherches sur la ligende sabine des engines de Rome (1967).

PRELIMINARY REMARKS

The tendency to make use of preexisting legends, epic or mythic,
in order to establish the beginnings of a history is certainly natural to
man, for one observes its effects whenever one reads such a history. I
shall cite only one example, remarkable because it is not yet a hun-
dred years old. It concerns another people with an Indo-European
language and tradition and bears on the same matter which at Rome
formed the legend of the two peoples. The Ossetes of the Caucasus,
sprung from the Alans, are the only existing descendants of the
Scythian family, a fact which in every respect gives them an impor-
tance out of keeping with their small numbers. Their language, which
is Iranian, is the last survivor of a vast group of dialects, and their
stories and popular rituals, which were not collected until the nine-
teenth century but are very ancient, often correspond with the
descriptions which Herodotus, Lucian, and Ammianus give of the
Scythians and the Sarmatae. In particular they have a collection of
epic legends, which has spread throughout all of the northern Cau-
casus, concerning the Narts, a fabulous people of ancient times. These
people consisted essentially of three families, each defined by a charac-
teristic quality, and the union of these qualities forms a structure which
is a visible continuation of the social classification — sometimes
theoretical, sometimes practical — common to all the Indo-Iranians,
into priests, warriors, and herdsmen-warriors: the Wise (Alsegatae),
the Strong (Aihsaertaegkatae), and the Rich (Boratse). A considerable
number of the stories are devoted to the bitter struggles of the Strong
and the Rich, with each family demonstrating its natural advantage:
on one side pure valor (sometimes assisted by great magic),
on the other opulence, trickery (sometimes dishonest) and great
numbers of men. The first Russian observer to record these stories
was seized with enthusiasm and by means of very bad plays on words
used them to construct the " prehistory ” of the Ossetes. Briefly,
according to him, the Boratae, a kind of Sabines, were the earliest
inhabitants of the country, which was later invaded by the Aihsaer-
taegkatae, foreigners coming from the north of Iran. The wars which
the epic texts relate were wars of installation and ended with a fusion;
after the death of the principal hero of the dEhsaertaegkatae his surviv-
ing companions settled down with their conquerors. V. B. Pfaff, who
made this reconstruction, was an important man. The first review of
Caucasian studies, which was in its fourth volume (1870), could not

FROM MYTHOLOGY TO HISTORY

reject his article, but the editors appended to it a flattering note in
which they disclaimed all responsibility. 19 This “historical operation”
was without a sequel. The nineteenth century was already too critical,
and another scholar, V. F. Miller, the true founder of Ossetic studies,
was already at work. The situation was not the same in Iceland at
(he end of the twelfth century when Snorri, humanizing the JEsir
and Vanir gods, used myth as the basis for “the earliest history”
of Scandinavia; not at Rome, three or four centuries before our era,
when respectable scholars filled in the gaps of time and space in
barium with the old traditional myth explaining how a society is
formed on the basis of three (or two) groups, each being the trustee
of one (or two) of the three functions necessary to its normal life.

The war of Romulus and Tatius thus presents an early example of
i he historidzation of myths, of the transposition of fables into events;
i his process was frequently used by the annalists or their predecessors,
and is characteristic even of Rome at this stage. During the past
quarter of a century a number of similar examples have been revealed
in the history of the royal period, as well as in the history of the first
war of the Republic. The structured antithesis of Romulus and Numa
defines two equally wholesome and necessary types of power, cor-
responding to those of the Vedic sovereign gods Varuna and Mitra. 20
I he whole story of the third reign, and especially the victory of the
third Horatius over the three Curiatii, reproduces the mythical
career of the warrior-god Indra, in particular the victory of the hero
Trita, “Third,” over the triple demon. 21 Consider further the two
mutilated men, whose conjunction at this stage of history is in itself
improbable. Codes and Scaevola, the Cyclops and the Left-hander,
successively save Rome when it is besieged by Porsenna, one paralyz-
ing the Etruscan army by the dazzling glance of his eye, the other
sacrificing his right hand before the Etruscan leader in a heroic act of
perjury. These two form a pair paralleling the one-eyed god and the
one-handed god of the Scandinavians, 6dinn and Tyr. The former of

iq. Sbornik svMenej o kavka^skih gorcah 4 (1870), pt. 1, sec. 3, pp. 1 (with n.i)-7, esp. pp. 4-5.
Sec rhe whole story in ME 1 : 545-49.

20. M V, chap. 2 ; summed up in ME 1 : 274-78.

21. Aspekte . . . chap. 1 (now also The Destiny of the Warrior [1970], chap. 1), where one
. .in find a criticism of the thesis (of H. J. Rose, etc.) which sees in the story of the Horatii
.hhI ihe Curiatii an ingenious fiction fabricated on the basis of the names of places or
preexisting objects (Tigi/lum Sonmwm, Pila Horatia, Janus Curiatius . . .); summed up in
\I/; 1 : 278-80.

76 PRELIMINARY REMARKS

these, because he has sacrificed an eye, receives supernatural wisdom
as compensation, while the other saves the gods by thrusting his
right hand into the jaws of the demon- wolf . 22

And so the information afforded by a consideration of the rites of
Mater Matuta is completed. There was indeed a Roman mythology,
just as rich as that of the Vedic Indians and the Scandinavians. To be
sure, large parts of it related to a great number of gods have disap-
peared. But other parts, and the most important ones — the three cases
which have just been cited concern the ideological provinces of Jupiter
and Mars — have survived. The myths have merely been transferred
from the world of the gods to the world of men, and their heroes are not
gods but the great men of Rome, who have assumed the characteristics
of those gods. In later times they continued, and perhaps even more
successfully, to act out the parts of exempla, providing incentives and
justifications, which is one of the functions of mythology.

To return to our point of departure, the comparative interpreta-
tion which has just been suggested by the account of Rome’s first
war eliminates one of the traditional explanations of the god Quirinus.
The meaning of this god, as we shall soon see, is one of the cardinal
points, indeed it is the hinge of the study in which we are engaged.
According to the way in which one conceives him, whether as a god
of the earliest Roman or even of a pre-Roman society, or as a god
imported and superadded by a foreign element, Sabine for example,
everything depends, and not merely in his sphere; Jupiter, Mars, and
many fertility gods are interdependent with him. We shall observe
him later at our leisure, but we can record this important negative
fact: the account of synoecism in the annalistic tradition, being of
mythical origin, cannot be regarded as a basis for the theory of a
Sabine Quirinus . 23

Despite the striking fidelity of important authors to the idea of a
primitive Sabine element , 24 it seems to be losing ground. Other

22. MV chap. 9 ("Le Borgne et le Manchot ”) ; ME 1 : 424-28.

23. A second consideration, the parallelism of the facts at Iguvium, will completely
eliminate this thesis ; see below, pp. 149-50.

24. This conviction readily replaces the proofs which it lacks with heated affirmation and
an appeal to authority. Thus C. Koch, Reiigio (i960), p. 25: "Jedenfalls gibt es heute keinen
Bodenforscher, der es wagte,” etc. The arguments of Jacques Heurgon in favor of a Sabine
component in the earliest population of Rome, Rome ct la Mediterranee orientate jusquaux

FROM MYTHOLOGY TO HISTORY

authors, no less important, have recently abandoned it. But even
among the latter group, when it comes to Quirinus, there is something
in the manner of explanation suggested by the annalistic tradition. The
< ollis Quirinalis, they say, whether settled by Sabines or Latins, whether
lacing the Palatine-Equiline region or being the advance outpost of
i his region, was in any case the hill of Quirinus and was named after
Quirinus. As this locality, lying outside the first grouping of the
montes, did not become part of the urbs until later, by a true synoe-
cism, its special god must have joined the common pantheon at this
stage. Except for nationality, the explanation of the presence of
Quirinus and of his role in Roman religion does not undergo any
change; he is always an originally separate god — attached, if not to a
separate people, then at least to an external place — who will only
secondarily be added to the gods of the montes. We shall discuss this
explanation later, but it must be shown here that it is neither self-
evident nor logically necessary. In fact it rests on an unexpressed
postulate — namely, that the hill was named Quirinalis before its
incorporation into the city, and that Quirinus was its special, epony-
mous god from the time of the earliest settlement. But things might
have occurred differently. The other hill in the northern sector,
die Capitol, never officially took the name “hill of Jupiter”;
yet in the course of history it became the cultic abode of this god, who
was certainly worshipped from the earliest times on the same level
as Mars. The connections between Quirinus and the terrain may have
been parallel, his installation on a new hill having, in addition, in-
volved a change of name. Later we shall see that there are other
reasons for thinking that an organic triad, Jupiter-Mars-Quirinus,
presided from the beginning over Latin society in its earliest habitat,
during the first regnum. This society having gradually annexed all the
bills, the worship of its principal divinities may have been distributed
,ii the end of this expansion, with Jupiter taking possession of the
( lapitol and Quirinus occupying the hill which thenceforth would
be named after him, Quirinalis. The choice between these opposed
explanations of the god, whether fundamentally as an Ortsgottheit
or as an imported god, will depend on what less external considera-
tions tell us about his nature and his function. At least we have
learned that the name of the collis Quirinalis does not necessitate the

78 PRELIMINARY REMARKS

This reflection, we may say in passing, applies not only to the collis
Quirinalis; there is no rigid, absolute bond between a cult and the
place where it was practiced during the historical era. Although the
Regia and the aedes Vestae as we know them are in the Forum, we must
guard against concluding that the conceptions, rules, and rites which
concern them are subsequent to the occupation of the Forum. From
the time of the earliest settlement there were surely a rex and a
Regia, as well as a public hearth. When the Forum had been occupied
under adequate conditions of security, the rex and the Vestals, the
royal house and the hearth, were installed there, bringing with them
as a matter of course their particular qualities and their traditional
singularities.

THE INDO-EUROPEAN
HERITAGE AT ROME

The preceding pages have given several examples of the necessity of
taking the Indo-European factor into account in the investigation and
understanding of the earliest Rome, that is, of comparing Roman
facts with the homologous facts of India or Scandinavia, or Ireland or
( )sset ia. Is this comparison possible? Some make th t a priori assertion
that it is not. Thus Kurt Latte, who is willing to excuse self-deception
in the romantics of the nineteenth century, but not in our era (p. 9):

The time which elapsed between the arrival of the Indo-Europeans in the
Mediterranean regions and our sources could well appear to be without
importance in this epoch [that of Max Muller and Preller], since Homer and
die Vedas were dated very early and the Indo-European invasion very late
(see the criticism of this opinion in Kretschmer, Einleitung in die Gesch. d.
X riech . Sprache , pp. 76 ff.). They failed to account for the changes which a few
« enturies can produce in the thought of civilized peoples, and for the cultural
mlluences which dominated the development of the Romans. Today we know
I low recent our tradition about Rome is.

To clever men who contended that motion was not possible, the
philosopher replied by walking toward them. We shall do the same
here on the numerous occasions when the comparison with India
or Scandinavia will clarify obscure points or exorcise the monsters
born from an uncontrolled criticism. But the objection can be
destroyed even in its principle.

The oldest Irish traditions are known only in the form which they
look in the high Middle Ages; the oldest Eddie poems date from the
hist centuries of the first millennium of our era, and the bulk of our
information comes from Snorri, who lived at the end of the twelfth

PRELIMINARY REMARKS

and the beginning of the thirteenth century. And yet the comparison
of these Western texts with those of the Vedic society, which was
older by two thousand years, has produced enlightening results
on both sides. I do not speak of my own work or that of my closest
friends. Two eminent specialists who were both trained in comparative
argumentation, the Celtologist Joseph Vendryes and the Indologist
Sylvain Levi, agree in admiring how not only the ri and the rdjan , the
druid and the brahman, but also the forms of epic and court poetry
in their respective fields resemble each other and how each con-
tributes to the better understanding of the other. Today Myles Dillon,
who is both an Indologist and a Celtologist, analyzes this idea in de-
tail, just as Alwyn and Brinley Rees in Wales do in a fine recent book.
Others have shown that the rhetorical character of certain poems of the
Edda parallels that of the Gathic poems, which are the oldest portions
of the Avesta . Was Rome, which was known earlier than the British
Isles or Iceland, truly so singular that the same kind of comparative
investigation cannot be applied to it?

Even more surprising, the religious vocabulary of Rome itself
provokes this investigation. In 1918, developing the suggestions of
Paul Kretschmer, Vendryes published an article of primary impor-
tance in the Memoires de la Societe de Linguistique de Paris, entitled “Les
correspondances de vocabulaire entre Tindo-iranien et l’italo-cel-
tique.” 1 Later scholars have generally dropped the term Italo-Celtic,
and the preferred wording today would be “ among the Indo-Iranian,
Italic, and Celtic languages” — some would even break down the
Italic unity — but this does nothing to alter the lesson of these twenty
pages, which the author summed up very well in the following
terms : “ What is striking is that a rather large number of words appear
in this list of concepts which are connected with religion, and especially
with the liturgy of worship, with sacrifice. Reviewing these words,
adding certain others to them, and grouping the whole by categories,
one does not merely establish one of the most ancient elements of
Italo-Celtic vocabulary; one also establishes the existence of common
religious traditions in the languages of India and Iran and in the two
Western languages.” Vendryes, strictly a linguist and a Voltairean
who quickly tired of the observation of religious facts, did not himself
measure the import of this eloquent statement. But the statement

1. 20:265-85.

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Texts

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Archaic Roman Religion"

See other formats